Civilian Radioactive Waste Management System

Management & Operating Contractor

THERMAL TEST PROGRESS REPORT #3

July, 1999

Prepared for:

U.S. Department of Energy Yucca Mountain Site Characterization Office

P.O. Box 30307 North Las Vegas, NV 89036-0307

Prepared by: Natural Environment Program Operations

Civilian Radioactive Waste Management System Management and Operating Contractor

1261 Town Center Drive Las Vegas, NV 89134-6352

Under Contract Number DE-AC08-91RW0034

EXECUTIVE SUMMARY

This Thermal Test Progress Report # 3 is the third of a series of informal reports intended to communicate the progress of the in-situ thermal tests. The progress reports are prepared and distributed every three months or so.

The Large Block Test (LBT) and the Drift Scale Test (DST) are the two components of the current in-situ thermal testing program at Yucca Mountain.

The Single Heater Test Final Report, a level 3 deliverable, has been submitted for review by DOE. The outcome of the Single Heater Test, as given in the final report, was covered in the previous Thermal Test Progress Report.

The Large Block Test is discussed in Section 2. The heating and cooling phases of the LBT are complete and measurements of temperature and moisture content of the block have been terminated. Overcoring for post-test characterization of the block has been completed. Laboratory testing, modeling and analysis will continue over the next few months.

The Drift Scale Test is covered in Section 3. The heating phase of the test is in its 2 0 'h month now. It is expected to extend approximately four years. Measurements of various kinds being made in the DST are presented and discussed in Section 3.

ii

CONTENTS

Page

1. INTRODUCTION 1-1

2. LARGE BLOCK TEST 2-1

3. DRIFT SCALE TEST 3-1

3.1 Power and Temperature Data 3-1 3.2 Thermo-Mechanical Measurements 3-17 3.3 Active and Passive Monitoring in the Hydrological Boreholes 3-19 3.4 Electrical Resistivity Monitoring 3-23 3.5 Ground Penetrating Radar 3-32 3.6 Neutron Logging Measurements 3-34 3.7 Isotopic Analyses of Gas and Water Samples 3-42 3.8 Water Sampling and Geochemistry Activities 3-45 3.9 Acoustic Emission/Microseismic Monitoring 3-53 3.10 Video Imaging Inside the Heated Drift 3-56 3.11 Heat Loss Through the DST Bulkhead 3-57 3.12 Interpretive Analysis of the Thermo-hydrologic Processes 3-61 3.13 Thermo-hydro-chemical Model 3-64

iii

1. INTRODUCTION

Two types of formal reports have generally been employed to date to present and discuss results of the in-situ thermal tests. These are the level 4 deliverable reports and the level 3 deliverable reports. The level 4 or the more numerous kind of reports cover one or a few aspects/subjects associated with a particular test such as the Single Heater Test or the Drift Scale Test. Usually, they cover all the work performed by a teammate organization with respect to a specific test. The level 3 reports, which are fewer, are intended to document in a comprehensive and integrated fashion all aspects of a particular test over a specific period of time. The contents of a level 4 report are incorporated in a subsequent level 3 report associated with the test. Both level 4 and level 3 reports are subject to rigorous guidelines of product quality in terms of reference citations, data presentations and interpretations, to ensure traceability, transparency and QA pedigree of information presented and discussed.

Because of the unavoidable overlap/duplication of efforts in preparing both the level 3 and level 4 reports and in order to streamline the process so that resource utilization is optimized, it has been decided to prepare only integrated level 3 reports for the thermal tests starting in FY99. There will still be some level 4 deliverables from each teammate organization comprising semi-annual data submittals to the technical database. However, there will be no formal level 4 reports. Elimination of the level 4 reports is not expected to adversely impact the communication within the thermal test team. Such communication is continuously ongoing and is greatly enhanced by the workshops every three months. Since the level 3 reports will be approximately 12 months apart, the absence of the more frequent level 4 reports may cause a gap in others' awareness about the progress of the thermal tests. To avoid such a situation, a series of informal reports, covering all aspects of the thermal tests, will be prepared and distributed. This report, designated Thermal Test Progress Report #3, is the third of the series of informal reports prepared shortly after each quarterly workshop to ensure wider dissemination of information on the progress of the thermal tests. Although, much of the information in the progress reports is derived from the preceding workshop, these reports are not limited to a summary of the workshop presentations and discussions. Any information, relevant to the in-situ thermal tests, available at the time of publication of a progress report is included and discussed in it.

This Thermal Test Progress Report #3 is prepared following the eighth workshop held at Las Vegas in April, 1999.

1-1

2. The Large Block Test

2.1. Introduction

The Yucca Mountain Site Characterization Project investigates the suitability of Yucca Mountain, Nevada as a potential repository for high-level nuclear waste. When emplaced in a repository, the radioactive decay heat of the waste may cause coupled thermal-mechanicalhydrological-chemical (TMHC) processes in the near field. These processes must be understood before model calculations can be performed to confidently predict the near-field environment, which will have deterministic effects on the performance of a repository.

The Large Block Test (LBT) is an integrated test with controlled boundary conditions so that results will be more useful for evaluating the coupled TMHC processes. Specifically, the LBT allows us to study the dominant heat transfer mechanism, condensate refluxing, relationship between boiling point isotherm and the drying of the rock mass, rewetting of the dry-out zone following the cool-down of the block, displacement in fractures, and rock-water interaction. The heating/cooling phases of the test were conducted from Feb. 1997 to Sept. 1998 and results have been presented at previous thermal test workshops. This presentation will cover the current status of the post-test characterization activities now underway for the LBT.

2.2. Description of the LBT

An outcrop area at Fran Ridge was selected to be the site for the LBT because of the suitable rock type exposed and accessibility of the site. A 3 x 3 x 4.5 m block of fractured nonlithophysal Topopah Spring tuff was isolated at Fran Ridge. Instruments and heaters were installed within and on the surface of the block. The instruments installed in the block included resistance temperature devices (RTD) to measure temperatures, electrodes to conduct electrical resistivity tomography (ERT), Teflon liners for the neutron logging in boreholes, Humicaps to measure relative humidity, pressure transducers to measure gas phase pressure, conventional and optical multiple point borehole extensometers (MPBX) for measuring displacements along boreholes, fracture gauges mounted across fractures on the block surface to monitor fracture deformation, Rapid Estimation of K and Alpha (REKA) probes to measure in-situ thermal conductivity and diffusivity, and visual observation of the drainage of water near the bottom of the block. Coupons of waste package material and introduced material were placed within the block to study the effect of the heated environment on the materials. Labeled local microbes were introduced back into the block to study their survivability and migration.

The temperature measurements included the spatial and temporal variation of the temperature in the block and the thermal gradient on the block surfaces. One heater of 450 Watts was installed in each of the five horizontal heater holes at about 2.75 m below the block top. A heat exchanger was installed to control the temperature on the block top. A layer of RTV and Viton were installed on the block sides to minimize moisture flux. Three layers of insulation materials (Ultratemp, fiberglass building insulation, and Reflectix) were installed on the outside of the moisture barrier. All of the instrument holes were sealed either by cement grout, packers, or an RTV/Teflon membrane. Straps were used to stabilize the block and insulation during the test. The heating phase of the Large Block Test began on February 28, 1997 and ended on March 10, 1998 when the heaters were turned off. The cooling of the block was monitored until

2-1

September 30, 1998 when it was determined that the block had returned to ambient temperature and the Data Acquisition System was turned off. During the heating phase the block was heated from within to reach a temperature of 140 *C at the heater horizon, and the heat exchanger was used to keep the top temperature at about 60 "C.

2.3. Post-test characterization activities.

Post-test characterization activities were started in early October 1998 and consist primarily of drilling/coring to provide boreholes and samples for post-test characterization of the block, and laboratory testing and analysis of core samples collected by the drilling/coring activity. In addition, numerical modeling of the hydrologic and geomechanical behavior of the LBT is also being conducted.

Drilling/coring activities.

The post-test drilling and coring activities include drilling of nineteen new boreholes and one overcore of an existing borehole. These holes are arranged as shown in Figure 2-1, which shows two vertical fans of boreholes drilled from the North and West sides of the block, respectively. Drilling of the post-test characterization holes started on Nov. 1, 1998 and was completed by February 19, 1999. All holes were successfully drilled, core recovery was high, generally greater than 90%. The core samples are archived at the Sample Management Facility. Video logs of the new boreholes were taken after the drilling was completed.

Laboratory testing/analysis of post-test samples.

A series of laboratory measurements, tests and analyses are underway to characterize the post-test core samples. These include the following.

-Physical properties measurements such as density, porosity, moisture content. A suite of samples representative of all of the portions of the block sampled by the post-test holes were selected by examining core at the SMF. These samples were shipped to LLNL, and measurements of density, porosity and moisture content are underway.

-Mechanical properties measurements including uniaxial compressive strength and Young's modulus. Tests to determine post-test values of these parameters will be performed on samples prepared from the overcore of heater hole #4. Test specimens have been prepared and the MT&E to be used are now being calibrated.

-X-ray Diffraction (XRD) analysis to determine mineral content of the matrix material. Samples for a limited XRD study of post-test core have been selected and prepared and analysis of the mineral content is underway.

-Examination of fracture surfaces using Scanning Electron Microscopy (SEM) to determine type and amount of secondary mineralization that may be associated with hydrothermal processes. Samples for a limited SEM analysis have been selected and prepared and analysis is underway.

2-2

2.4. Analysis and Modeling

Fracture Analysis.

The LBT was conducted in a fractured rock mass, and as part of the block characterization activities, fractures in the rock mass were mapped and characterized. A three dimensional model of fractures in the block was developed based on data for fractures mapped on the surface of the block and fractures mapped in the interior of the block as derived from video logs of the boreholes drilled into the block (Section 2.2, Wilder et al, 1998).

During this reporting period video logs of the post-test boreholes (shown in Figure 2-1) were analyzed to obtain data on fractures intersected by these holes. These data were added to the fracture data base. The fracture data were then analyzed and the major through-going structures that penetrate the LBT were identified and modeled. This involved correlating the surface fracture traces with the location of fractures intersecting the boreholes. Correlation of the fractures mapped in the boreholes with the surface fractures was confirmed using the location, strike and dip of the fracture as measured in the video log. Figure 2-2 presents a series of 3-D perspectives of the major fractures cutting the block.

The analysis shows that the fractures in the Large Block Test can be grouped into six fracture systems based on similarity in strike and dip as shown in Figure 2-3. These results along with detailed descriptions of each fracture set have been described in a report that is currently in review (Wagoner, 1999).

Geomechanics Modeling.

A discrete element model of the Large Block Test was developed using the 3DEC code (Itasca, 1998). In this model the rock mass comprising the Large Block Test is represented as an assemblage of discrete rock blocks separated by fractures as shown in Figure 2-4. The model uses a subset of the mapped fractures as input. The fractures chosen for this model are thought to be the most important fractures in the block, and also include the fractures monitored for displacement using 3-component fracture deformation instrumentation (Section 4, Wilder et al., 1998). Preliminary results are consistent with observed displacements and indicate opening of a major vertical E-W trending fracture above the heater plane after 70 days of heating (Figure 2-5). This is consistent with the development of a refluxing zone in this region of the block.

2.5. Large Block Test Summary

Afracture model of the LBT has been completed and documented in a report. A discrete element model for simulation of deformation during the thermal cycle of the test has been developed, and preliminary results are consistent with deformations observed during the test. Coring of the block has been completed, samples prepared, and experiments and studies are underway to determine the response of the rock to the hydrothermal environment. Results of the post-test characterization activities will be used with data gathered during the test to enhance our understanding of the coupled TMHC processes. Thermal-hydrological, thermal-mechanical, thermal-chemical, and the coupled processes models will be constructed to compare with the

2-3

observed results. Through the comparison of the observations and the model calculations, the coupled TMHC processes will be better understood. The processes may be included in models used to predict characteristics of the near-field environment of a repository.

2.6. References:

ItascaConsulting Group, Inc "3DEC, 3 Dimensional Distinct Element Code - User's Guide", Itasca Consulting Group, Thresher Square, Minneapolis, MN, 1998.

Wagoner, J. L. "Fracture Characterization of the Large-Block Test, Fran Ridge, Yucca Mountain, Nevada" UCRL-ID-133846, Lawrence Livermore National Laboratories, Livermore CA, 1999.

Wilder, D. G., W. Lin, S. C. Blair, T. Buscheck, R. C. Carlson, K. Lee, A. Meike, A. L. Ramirez, J. L. Wagoner, and J. Wang, "Large Block Test Status Report", UCRL-ID-128776, Lawrence Livermore National Laboratories, Livermore CA, 1998.

Acknowledgment Work performed under the auspices of the US. Department of Energy by Lawrence

Livermore National Laboratory under contract W-7405-ENG-48. This work is supported by Yucca Mountain Site Characterization Project, LLNL

2-4

Figure 2-1. Schematic locations of post-test holes drilled in the Large Block Test as viewed from the North West corner of the block. All holes were drilled dry and were continuously cored.

2-5

Figure 2-2. Perspective depiction of fractures intersecting the large block.

2-6

Figure 2-3. Equal area projection of major fractures identified in the large block rock mass.

2-7

Biocks !and Grid for 3DEC Analysis

NorthEast

Figure 2-4. Discrete Element representation of the Large Block Test

2-8

Figure 2-5. East-West cross section of discrete element model showing predicted displacement after 75 days of heating. Note: opening of vertical fracture is indicated above the heater plane. This could create a path for hydrothermal refluxing observed in the test.

2-9

3.0 Drift Scale Test

The following sections describe different aspects of the Drift Scale Test.

3.1 Power and Temperature Data

This section describes the thermal data obtained from the Drift Scale Test between its inception and March 31, 1999. The data that will be described include the heater power data and the temperature data obtained from the heaters, the interior of the drift, and the rock near the drift.

Power Data

Figure 3.1-1 illustrates the power being supplied to all of the wing heaters and all of the canister heaters. These total power data are Q measurements. The canister heater power decreased from about 53 kW at the beginning of the test to about 51 kw on day 244 (8/4/1998), a drop of about 2%. From day 244 to 270 (8/30/1998) the power increased to about 53.5 kW. This increase is likely due to changes in the ventilation system just outside the Heated Drift, which were implemented on 8/4/1998. These changes resulted in increased airflow near the cables supplying power to the canister heaters. This increased airflow likely cooled the cables, thereby decreasing the cable resistance. Decreased resistance increased the current flow in the heater circuit, thereby increasing the power to the heaters.

The power being supplied to each of the nine individual canister heaters are illustrated in Figure 3.1-2. These individual canister heater power data are not Q measurements. The black curve in Figure 3.1-2 is the total power being supplied to all the canisters divided by 9. This should be the average canister power but in fact exceeds all the unqualified individual measurements. The total power measurement is the more defensible measurement of the canister heater power.

The power distribution to individual wing heater elements on day 483 (3/31/1999) is illustrated in Figure 3.1-3. Of the 100 wingheater elements, 4 have failed and no longer supply heat to the rock at all The outer element of wingheater 9 failed after 79 days of heating, both elements of wingheater 29 failed after 185 days of heating and the outer element of wingheater 26 failed after 211 days of heating. Three elements are supplying significantly less than the average power (the outer element of wingheater 7, the outer element of wingheater 20 and the inner element of wingheater 25). Two elements are supplying significantly more than the average power (the outer element of wingheater 8 and the inner element of wingheater 17). While these anomalous power levels effect the rock temperatures in the immediate vicinity of the respective heaters, the overall impact on the test is minimal.

3-1

Thermal Data from the Heated Drift

Figures 3.1-4, 3.1-5, and 3.1-6 illustrate the temperatures on the canister heaters, in

the air in the Heated Drift and on the walls of the Heated Drift, respectively. By day

483 (3/31/1999) the canisters had reached temperatures as high as 184 'C while the

air and wall surface temperatures were in the 150 to 170 °C range. For each of these

threedata sets, the temperatures started out at about 30 'C just prior to the start of the

test. When the heaters were activated, the temperatures rose very rapidly at first, but

then the rate of increase of temperature decreased with time up until about day 130 (4/12/1998). From day 130 to day 483 (3/31/1999) the temperatures increased nearly linearly. The reason the rate of temperature increase stabilized is likely due to the presence of the wing heaters; without them, the rate of temperature increase would likely have continued to decrease. Linear extrapolation of the drift wall data suggests that the drift wall temperature will reach 200 °C on approximately day 700 (11/1/1999).

On day 239, there was a drop in the canister, air and surface temperatures of about 5 'C. This resulted from changes to the air ventilation system on the cool side of the bulkhead, which caused more cool air to be circulated onto the surface of the bulkhead. For approximately 5 days, cool, ambient air was forced into the Heated Drift through a hole in the bulkhead. When the hole was closed, the temperatures started to increase again. It is interesting to note that after the hole was sealed the rate of temperature increase, while still nearly constant, appears to be slightly less than the rate of temperature increase prior to the incident with the hole. This likely reflects an increase in heat flux out of the bulkhead as a result of the increase in the vigor of the ventilation system on the cool side of the bulkhead. On about day 309 (10/8/1998) the ventilation system on the cool side of the bulkhead was modified in a manner that reduced the airflow near the bulkhead. This change resulted in less heat flux through the bulkhead and the canister, air and rock wall temperatures all increased slightly but noticeably. The 12 hour long power outage on day 420 (1/27/1999) resulted in noticeable temperature drops in the canister, air and drift wall temperatures, but they all recovered quickly after restoration of power.

Figure 3.1-7 illustrates the air temperature in the heated drift after 483 days of heating, as a function of position in the drift and Figure 3.1-8 is a contour map of the drift wall temperature on that same day. These data indicate that the middle part of the drift was about 10 °C warmer than the ends at that time.

Figure 3.1-9 illustrates the air pressure and humidity' measured inside the Heated Drift. These measurements are correlated. When the air pressure increases, relatively dry air from outside the Heated Drift is forced into the drift decreasing the humidity. During the time that outside air was being forced into the drift through the hole in the

bulkhead there was a pronounced decrease in humidity in the Heated Drift, but no appreciable change in air pressure,. There has been an overall decrease in the relative humidity inside the heated drift since the beginning of the test. Since the temperature

3-2

inside the drift has also increased substantially, this does not necessarily indicate that

the air has lower water content.

Figure 3.1-10 illustrates the temperatures recorded by the temperature sensors mounted on the hot side of the thermal bulkhead. The pronounced changes in the temperature distribution on the bulkhead were caused by changes in the thermal insulation on the outside of the bulkhead and in the ventilation system just outside the drift.

Rock Temperature

Figure 3. 1-11 illustrates the locations of the 28 boreholes that contain RTD temperature sensors grouted into them.

Figure 3.1-12 shows the temperatures observed after 483 days of heating (3/3111999) in the four horizontal boreholes emanating from the Heated Drift. The data indicate that the rock adjacent to the wing heaters is warmer than the rock near the Heated Drift. This is because of the higher power output of the wing heaters as compared to the canister heaters in the drift. The temperatures near the wing heaters show two pronounced "humps", which correspond to the locations of the two wing heater elements deployed in each borehole. The data from the left side of the drift at Y=12 meters do not exhibit a noticeable "hump" associated with the outer wing heater since this hole is located midway between wing heaters 6 and 7 and the outer heating element of wing heater 7 is operating at only about 75% power. Three of the four temperature profiles exhibit a slight shoulder about 15 meters from the drift at a temperature corresponding to the boiling point of water. These are manifestations of a region of rock where water in the pores of the rock is boiling.

Figure 3.1-13 shows the temperatures measured in borehole 173, which is located at Y = 39 meters and which extends vertically downward from heated drift. Note the isothermal region which extends from about 2 to 4 meters from the drift wall where the temperature is approximately constant at about 97 0C. In this region, water in pores of the rock is boiling and liquid water and steam coexist in the rock.

3-3

QA Wing Heater and Canister Power Levels -

140

135

130

125

60

55

50

100 2Z00 300 400 500Time (days from heater activation)

Figure 3.1-1 Total power measurements (Q) for Wing Heaters and Canister Heaters.

3-4

a.)

0

a.>

0

Canister Heaters I

I

Canister Heater Power

100 200 300 400 500

Time (days from heater activation)

Figure 3.1-2 Power measurements (not Q) for each of the Canister Heaters

3-5

6.2

6

5.8

5.6

5.4

5.2

0

5

4.8

4.6 L0

Wing Heater Power After 494 Days of Heating (3/31/1999)

45

40

35

mem

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

*42

43

44

45

46

47

48

49

50

0 1 2 0 1 2

Power(kW) Power (kW)

Figure 3.1-3 Power distribution (not Q) to wing heater elements on day 494.

3-6

MIENNE

30

25

20

15

C C

Q

10

5

•III Il••• -II

,,dl

Canister Thermocouples

200

1 8 0 . . .:.. ... . ............................... ........... .......... : .. . .. ....... ...............................

160

120 0.10.200 4 1 2 - ..... ............. . .... .... ........... .. ............ ..... ..............................

8 0 .. ... .. .. .... ... .... .. .. ..

.6. .0. .... ....................................... ...................................................

4 0 . ..... ......... .......................................................................... .................. ............ .

20 0 100 200 300 40q

Time (days from heater activation)

Figure 3.1-4 Temperatures on the canister heaters.

3-7

0

0)

S4

HD

0 500

Air Temperature in the Heated Drift

- 0 100 200 300 400

Time (days from heater activation)

Figure 3.1-5 Tempeeratures on the air in the Heated Drift.

3-8

180

160

140

120

100

80

&

0 01

60

40

9I500

•v

Drift Wall Temperature

0 100 200 300 400

Time (days from heater activation.)

Figure 3.1-6 Temperatures on the walls of the Heated Drift.

3-9

180

160

140

120

100

80

0

S H

60

40

20500

Air Temperature in the Heated Drift After 483 Days of Heating

170

165

160

155

150' C 10 20 30 40

Y (meters)

Figure 3.1-7 Temperatures on the walls of the Heated Drift on Day 483.

3-10

U 0

Sý

I I I ~ I I I II I I I

AL L A

. ............................. .. ................................ ................................. ................................. .. ............................

0 Crown - Right Rib A Left Rib

I I , i , i , , I ,

50

45

40

35

30

25

20

15

10

5

0

K

-5 Distance

Day 483 Tavg 161.3 'C

6 5 4 3 2 1

0 -1 -2 -3 -4 -5 -6 -7 -8 -9

0

0 4

0 5

from Roof (m)

Figure 3.1-8 Contoured temperatures of the wall of the Heated Drift on Day 483.

3-11

Air Pressure and Humidity in the Heated Drift

100 200 300 400

60

50

40

30

20

501

Time (days from heater activation)

Figure 3.1-9 Relative Humidity and air pressure inside the Heated Drift.

3-12

13

12.5

12

11.5

Bt

10

0"L 0

Temperatures on the Hot Side of the Bulkhead130

1 2 0 . .. .. .: ... .... .. .... .... .... .. .... . .... . .. .......... ..:. .... ........ .. ... ........... .... ... . .. ........ ... .... ..... .... .. .. .. .. .. I ... .

1 1 __ -. . .:. . . . . . . . . . . . . . ......................... .. ............. . ..... .. ..... ........... ............. ..... 1 1 0 ...... .......................

1 0 _ . ... ... .. ......... .. .. ....... .... .. .. .. ...... . .... ... ... .. .. .. .. ... ... ..... .................. 100 X .... .. ... ... ............. ..... .. .... ............. .. ...... ........

S 9 0 . ................ ......... 0

7 - _ . . . .......... .. .... ................. S/8

c~70- ...

S6 0 ...........

E--i 50

4 0.,•... .... ............ i S

40

20. .. ...~~,

0 100 200 300 400 500

Time (days from heater activation)

Figure 3.1-10 Temperatures measured on the hot side of the Bulkhead.

3-13

Y=3 Mete rs

133

80 79 m mN•

imm~l

134

Y 9 = M444n rcz

80

Y=39 m

Y=32 m

Y=23 m

Y=12 m

Y=3 m

79

+T:

168

80 79

169

Y-39 Meterm

171 .170

175

80. .79 mllll

mlmm

12174

.173

I- RTD designed locations - RTD as-built locations - Wing heater as-built locations

Figure 3.1-11 Schematic locations of RTD arrays and wing heaters.

3-14

Temperatures in Horizontal Boreholes After 483 Days of Heating

-20 -15 -10 -5 0 5 10 15 20 25

X (meters) Figure 3.1-12 Temperatures in four boreholes emanating from the Heated Drift.

3-15

220

200

180

160

140

120

100

& 0

d.)

S H

_,,,Ii , ,II l l I•tI I l1111 1111 11I 111 I V [I FTl F •

- .. . . ..... i................ ............ . : ...... ... .. ................ : ................ : .............. -............... :.... D a 4 3

_: ........... ............... ..........................i' ...... ........................................i .......................................... : /

S............ ................ ... .............: ........ .. ... .......

- N

............... ....... ..... .. .. .... .... .. . .. .....

-" ........... i .......... .... ...... .......... . . . . . . . ................ . ................................ ... ............ "......

_- / ,

_ ............ .... ........ ................. ,........... t • • ........ ............. ............ . ......... .RTD139 .. RTD143

. ..... RTD160 S. ..' ..................... ............ t • ........... : ....... ....: ......... ...... .• ...... - " ",RTD164I ..

80

60

40

202 -25

Temperature in RTD-173 After 483 Days of Heating

2 4 6 8 10 '12 14 16 18 20

Distance from Borehole Collar (m)

Figure 3.1-13 Temperatures in borehole RTD-173 after 483 days of heating.

3-16

150

140

130

120

110

100

90

80

70

60

& 0

0

0

S 0 H

50

40

30

201

3.2 Therino-Mechanical Measurements

This section discusses events and trends associated with thermal and mechanical data from 1/1/1999 through 3/31/1999. The discussion here follows on to that of the previous data in the SNL report "Drift Scale Test Draft Status Report #1: Evaluation and Comparative Analysis of the Drift Scale Test Thermal and Thermomechanical Data (Results of 12/3/1997 Through 5/31/1998)" dated July 15, 1998.

Thermal Data

The maximum temperature in the Heated Drift and on its walls (disregarding surface thermocouple SURF-TC-39, which has been anomalous since the initiation of the heaters) as of 3/31/1999 (452 days of heating) was measured to be 1630C. A new anomaly was seen in the temperatures in the MPBX boreholes 81 and 82. Temperatures along the length of these boreholes suddenly jumped, one by one over a period of several days, to the local boiling temperature of 960C. The jumps occurred at approximately 1116m from the collars, which would put them near the Y coordinates 0-5m, or near the bulkhead location. There is a slight indication of increased expansion in the MPBXs, but not as dramatic as the temperature increases on the thermocouples in the open boreholes might indicate.

Mechanical Data

The behavior exhibited thus far by the MPBXs continues to be similar to the elastic model predictions, with the exception of MPBX-14, which demonstrates separation between the invert and liner. MPBX-14 indicates de-bonding of the invert from the liner, with a separation as much as 7 mm. A movement in MPBX-14 at 230 days possibly corresponds to a change in the vertical cross-drift extensometer at about the same time. The deepest anchor (Anchor #4) for the MPBXs is progressing toward having the largest displacement from the collar. The inconsistent behavior of the MPBX sensors (LVDTs) is difficult to understand; some of the LVDTs for MPBX-7 and -11 have apparently begun operating again, MPBX-3 has lost its noise problems, and MPBX-8, -12, and -14 have gotten progressively noisier during the 5h quarter.

The data shows little evidence of non-elastic behavior in terms of specific events or reversals of trends. Possible slips for Anchors #1 and #2 are seen for several MPBXs (#6, 9, 11, 13); however, the noise associated with these events make it difficult to determine

.whether the data are expressing events in the rock. The magnitudes of displacement are somewhat higher (as much as 40%) than those predicted by elastic analyses using intact rock mechanical properties (E=37 GPa, SHT intact rock thermal expansion coefficients), particularly for the closest anchors (1 and 2). This may be indicative of joint deformation behavior near the drift wall, perhaps due to enhanced fracturing near the surface due to mining. A similar phenomenon was observed for the Single Heater Test when comparing elastic and compliant joint predictions to the test data; the compliant joint model predicted higher displacements that were more consistent with the data. SNL is initiating new compliant joint analyses to evaluate the DST MPBX data.

3-17

The horizontal cross-drift extensometer show a fairly constant closure distance of 5.5m over the most recent 30-day period. The vertical cross-drift extensometer has gotten much noisier over the same time period, with an apparent reversal in trend from closure to expansion.

The strain gages placed on the concrete liner and on unconstrained concrete samples in the Heated Drift continue to show the combined effects of thermal expansion, dehydration-induced shrinkage, and mechanical stress imposed by the interaction of the concrete with the heated rock surrounding the drift. The results from the strain gages on the unconstrained samples exhibit behavior indicative of drying shrinkage due to dehydration, a phenomenon seen elsewhere in engineering literature. All of the strain gages on the liner surface are in extension by the end of the 4e quarter due to combined thermal and mechanical effects. The mechanical component of the circumferential strain gages on the liner consistently show that the crown of the liner is in compression while the rest of the liner experiences smaller magnitudes of compression and tension.

Thermal expansion coefficients for T>96°C have been estimated for the unconstrained coupons. Three of the coupons have a thermal expansion coefficient of -12 pstrain/0 C, whereas the fourth is at - 18-21 pstrain/°C. There is no apparent correlation of thermal expansion behavior to the type of concrete. There is a possible correlation of behavior due to placement of samples, based on the near linear nature of two of the samples versus a more curved nature to the others. However, documentation of the relative placement of these samples with each other and with the nearby canister heaters is insufficient to determine such a relationship. There is evidence from the axial strain gage data of potential bending due to asymmetric heating of end of drift (WHs 25, 26, 29).

3-18

3.3 Active and Passive Monitoring in the Hydrological Boreholes

The hydrological boreholes consist of three fans of boreholes, referred to as boreholes 57

to 61, 74 to 78, 185 and 186. These twelve boreholes have 46 packers installed. Intervals are referenced by the borehole number followed by the interval number, which ascends from the collar of the borehole to the back of the borehole. Packer string 186 was repaired in April after two packers deflated (186-3 and 186-4) leaving packer 77-3 as the

only packer not inflated in the hydrology boreholes. The 45 inflated packers section the twelve boreholes into 45 isolated zones for passive monitoring and active testing.

Passive Monitoring:

Passive monitoring of temperature and pressure in each isolated interval is continuing. Four zones show temperatures that are either at or above the boiling point of water to date: 60-3, 60-4, 77-3, and 78-3. Both borehole 60 and 77 pass close to the wing heaters, and below the heated drift. Passive monitoring shows close tracking between the barometric signal and the pressure fluctuations recorded within each zone. The zones located further from the borehole collars show greater damping of the barometric signal as would be expected. Recalibrated pressure sensors were installed in March 23 and 24, 1999. Offsets in pressure can be seen in the acquired pressure data and are attributed to differences in the calibrations of the replaced sensors. Since pressure data is used in a differential mode to calculate air-permeability, no effect on data collection and processing results from the installation of the new sensors.

Borehole 60-3 continues to show cyclical variations in pressure, which is the result of water filling the bottom of the isolated zone above the level of the pressure sensing port. Samples of water continue to be periodically collected from Borehole 60-3. Borehole 602, which previously showed cyclical variations in pressure, now follows the barometric trend, and no longer produces water during periodic sampling.

Active Testing:

In February 1999 and April 1999 constant mass flux air-permeability tests were conducted in all isolated intervals. Air was injected in each zone for one hour to allow the pressure field to approach a steady-state response and the transient recovery was subsequently monitored. Air-permeability is used to monitor moisture redistribution since increases in liquid saturation will show up as decreases in air-permeability.

The steady-state pressure response was analyzed using the same ellipsoidal flow equation that was used to process previously acquired air-permeability data sets. Figure 3.3-1 shows a comparison of air-permeabilities as a ratio to pre-heat baseline values for zones in boreholes 57, 58 and 59, which are all located above the heater plane. Figure 3.3-2 shows air-permeabilities compared with baseline values for boreholes 60 and 61, which are located below the heater plane. As a general trend, intervals located below the wing heaters and the Heated Drift show the greatest decrease in air-permeability, with zones

3-19

located immediately above the heaters also showing a steady decrease in air-permeability. Zones that are located further away tend to show either no trend or in a few cases, a slight increasing trend.

Evidence of strong increases in saturation, noted by decreases in air-permeability, has been observed below the heated drift in zones 60-2, 60-3, 60-4, 61-2, and 78-2. Above the heated drift plane strong increases in saturation have been noted in zones 76-3, 76-4, 58-2 and 59-2. In addition, we are beginning to see the effects of formation dryout in a few intervals that show recent increases in permeability after showing steady decreasing air-permeability trends during the first year of heating. Zones 59-2, 60-3 and 76-2 all display increases in air-permeability during the April 1999 series of tests.

3-20

Boreholes 57-59

2 Si • I I I

I I I f 1.8 ---------- -- -- - - - - -

1.6 I°" -J p57-I 1.6 -- - -- - .. . . .-- ....... -- -- - ......- "- -- ,- - - ---- .. . .57-2I I X •, I 57-3

1 -'- "57-4

= 1.2---------- --- ----------------- ---------.------------ - -58-1

--1 - - - - - " --- - • +58-23

' -- -- e59-1

II II 0.84 --- --- - ---- 0.4 ----------.-- -- . . 00 .4- -- -- - - - - - - -- ,-- - . . . . . . . . , -- --------. . . .

II I I I

10/19/1997 1/27/1998 5/7/1998 8/15/1998 11/23/1998 3/3/1999 6/11/1999

Date

Figure 3.3-1. Many of the isolated intervals above the heated plane show decreases in airpermeability, indicating increasing liquid saturation. Zones in the uppermost borehole, 57, show either an increasing permeability trend or no significant changes over time.

3-21

Borhaoes 60-61

2

1.8

1.6

1.4

1.2

1

0.8:

0.6

0.4

0.2

0

10/19 5/7/1998 8115/1998 11/29/1998

Date

Figure 3.3-2. Isolated intervals in the 60-61 boreholes, located below the heaters, have shown consistent decreases in air-permeability as heating continues, indicating increasing liquid saturation.

3-22

a

II II I II II I I

I ' - --- - - -"

II I I

SI I I

I I I I I I SII I _I

± -,- ---

TT

I x

I I

*60-1

160-2

60-3

X 60-4

+ 61-1

061-2

"-61-3

061-4

(1997 1/27/1998 3/3/1999 6/11/1999

3.4 Electrical Resistivity Monitoring

Introduction

This section describes electrical resistance tomography (ERT) surveys made at the Drift Scale Test (DST), between January and March, 1999. ERT is one of the several thermal, mechanical, and hydrologic measurements being used to monitor the rock-mass response during the DST. The purpose of this work is to map the changes in moisture content caused by the heating of the rock mass water, with a special interest in the movement of condensate out of the system..

The purpose of this section is to provide a brief synopsis of the progress made during the time period of interest. A comprehensive discussion of the process followed to produce the results shown here, equipment used, and assumptions made can be found in Wagner, 1998, pages 7-13 to 7-21.

Changes in Electrical Resistivity:

Tomographs of resistivity change corresponding to the time period between January 13, 1999, and March 9, 1999, are shown in Figures 1, and 2. The top part of each of these figures shows the tomographs collected along a cross-section parallel to the Heated Drift (HD). The lower left portion of the figures shows tomographs corresponding to a vertical plane intersecting the HID at right angles, and about 5 meters in from the bulkhead; we will refer to this plane as AOD 1. The lower right portion of the figures corresponds to a second vertical plane intersecting the HD at right angles, about 24 meters in from the bulkhead; we will refer to this plane as AOD 5.

The upper part of Figure 3.4-1 shows the changes in resistivity along the HD, calculated after 406 days of heating. Over half of the tomographs area is yellow-green in color, thereby indicating resistivity ratios near 1.0 (no change relative to the pre-heat case). Near the walls of the HD, the tomographs show resistivity ratios less than 1.0, thereby indicating that the resistivity has decreased relative to the pre-heating condition. After 461 days of heating(upper portion of Figure 3.4-2), most of the rock still shows a ratio of 1.0. The rock below invert of the HD shows resistivity ratios near or above one, thereby indicating that the resistivity is increasing relative to the baseline. These increases become slightly stronger as heating time. Near the edge of the concrete liner, the tomographs above and below the HD show resistivity structure with undulations that are roughly coincident with the edge of the concrete liner.

The images showing resistivity changes below the invert of the HD and parallel to its axis indicate a region of anomalous resistivity decrease. This anomalous region appears as a vertical "finger" located about 17 m from the bulkhead. The resistivity decreases associated with this "finger" become stronger with time (i.e., resistivity ratio gets smaller). Also note that, a hole develops within the "finger" region; this hole can be clearly observed in Figure 3.4-1. The rock within this whole showed strong resistivity decreases during the first several months of heating and now the resistivity decreases are

3-23

becoming weaker as the rock becomes increasingly drier. The root cause of this behavior is described in Blair et al (1998), pages 2-6 and 2-7, and in Figure 2-4.

The lower left portions of Figures 3.4-1, and 3.4-2 show the tomographs corresponding to AOD 1. Although the mesh consists of a large region around the electrode arrays, only the region inside the ERT electrode array is shown in the figure because the region outside the array is poorly constrained by the data. The region inside the HD is also masked because the technique does not measure rock properties in the excavated region. The tomograph sequence shows a region of resistivity decrease near the walls of the HD. The resistivity decreases become stronger with time and they extend farther into the rock above the ND. The region corresponding to the wing heaters location (nine o'clock position relative to the HD) shows relatively weak resistivity decreases.

The lower right portions of Figures 1, and 2, depicts resistivity-change tomographs sampling the rock mass along AOD 5, which is a second vertical plane that intersects the HD at right angles near its middle. The images show increases (ratios greater than 1.0 near the location of the wing heaters). Resistivity decreases are observed near the crown and invert of the HD. We suggest that the resistivity changes observed through March, 1998, are caused by temperature increases as well as saturation decreases. The AOD 5 results in Figures 1 and 2 make it clear that resistivity increases are observed in the rock closest to the wing heaters while resistivity decreases are observed farther away. This behavior is caused by the competing effects that saturation decreases and temperature increases have on electrical resistivity.

There are both similarities and differences between the AOD 1 and AOD 5 planes. Both show resistivity decreases, near the walls of the HD, that extend deeper into the rock above the crown of the HD than below its invert. A significant difference between the AOD 1 and AOD 5 results is observed in the vicinity of the wing heaters were the AOD 1 tomographs show weaker decreases in resistivity near the bulkhead while the AOD 5 tomographs show resistivity increases.

Most of the rock above and below the HD shows resistivity decreases (ratios below 1.0). As indicated in Blair et al., 1998, when temperature has increased and relatively little drying has occurred, the resistivity will decrease because the temperature increase has enhanced the mobility of the ions in the pore water. As significant drying occurs and the saturation is higher than about 30%, the pore water paths through which electrical charge moves become smaller, thereby tending to increase the bulk resistivity. However, the net change in bulk resistivity is a decrease relative to baseline because the increases in ion mobility more than compensate for the volumetric reduction of the pore water pathways.

The rock closest to the wing heaters in plane AOD 5, and the rock below the invert of the HD show resistivity increases (ratios above 1.0). As the temperature continues to increase and the saturation drops below 30%, the pore water paths through which electrical charge moves become increasingly discontinuous. This makes the bulk resistivity increase exponentially even though the increases in ion mobility caused by temperature are still

3-24

present. Thus, the net effect on the bulk resistivity is an increase above baseline (ie., the ratio becomes greater than 1.0) in hot dry rock with saturations less that 30%.

Estimates of Saturation Change:

The resistivity changes shown in Figures 1, and 2, influenced by changes in moisture content, temperature, and ionic strength of the water. To estimate saturation, it is assumed that the dominant factors affecting resistivity changes are temperature and saturation. That is, an increase in temperature or water saturation causes a resistivity decrease. Near the heaters, there may be regions where the increasing temperature reduces the resistivity, while rock drying changes the resistivity in the opposite sense (increases the resistivity). Our goal in this section is to use the images of resistivity change near the HD, along with the measured temperature field and what is known of initial conditions in the rock mass, to estimate moisture change during heating. A detailed description of this approach can be found on a previous report pertaining to the Single Heater Test (Blair et al., 1998, pages 2-5 to 2-7).

To estimate moisture content changes, the effects of both rock temperatures and resistivity changes measured by ERT must be taken into account. Interpolation of temperature measurements made at discrete points was necessary to develop the temperature fields showing temperature values at each tomograph pixel. The interpolations can sometimes be in error when the interpolation algorithm is not sufficiently constrained by the measured values and because the interpolation does not necessarily satisfy physical laws. Note that the temperature estimates were generated following a nonqualified process; therefore, the saturation estimates are nonqualified. Also, the saturations estimates are considered approximate, as indicated later in this report.

Figures 3, and 4 show the saturation change estimates corresponding to heating days 406 and 461, respectively. The results shown were calculated using Model 2 (assumes that the primary conduction pathway is the electrical double layer) as described in Blair et al., 1998, pages 2-5 to 2-7. Model 1 results (assumes that the primary conduction pathway is the electrolyte located away from the pore walls) for day 461 results are shown Figure 4.4-5.

Note that the Model 2 estimates on Figure 3.4-3 indicate that most of the rock immediately adjacent to the HD and to the wing heaters show saturation ratios below 1.0 These estimates suggest that some drying is occurring close to the crown and invert of the HD. A lot more drying is occurring close to the wing heaters. Also note that the drying zone is growing very slowly with time. The tomographs suggest that the drying zone extends farther into the rock above the HD and than below it. The saturation ratio estimates shown in Figure 3.4-5 suggest that the rock showing the maximum amount of drying is located in plane AOD 5, forming a horizontal zone centered on the wing heaters; also, the amount of drying higher near the AOD 5 wing heaters appears to be larger that near the AOD 1 wing heaters. The tomographs show that the rock near the AOD 5 wing heaters has lost more than four-fifths of its water (saturation ratio less than

3-25

0.20) by early March, 1999. When compared to AOD 1, AOD 5 shows a stronger zone of drying that extends beyond the zone where the wing heaters are located.

Saturations ratios greater than 1.0 indicate regions in the rock were water saturation is increasing relative to the pre-heating conditions; we will refer to these zones as wetting zones. Wetting zones are observed in Figures 3, and 4, such as those that occur below the HD near the bulkhead, as well as in discrete zones closer to the middle of the HD. In particular, the strongest wetting zone below the HD appears to be related to the "finger" resistivity anomaly associated with figures 1, and 2. Note that the "finger" resistivity anomaly correlates with a region of drying close to the HD invert, and to a region of wetting farther away from the invert. Other wetting zones appear above and near the outer edge of the wing heater location, within planes AOD 1 and AOD 5. Another wetting zone is present below and near the edge of the wing heaters in AOD 1.

Drying and wetting zones can also be observed in Figure 3.4-5; these estimates are based on Model 1. As indicated in Blair et al., 1998, the saturation estimates calculated depend upon the model assumed. Our best guess is that the model 2 results are closer to reality than Model 1 given the baseline rock condition. As heating affected the rock, it is quite possible that the Model 1 results are a better indicator of saturation. Thus, a wise approach is to consider the results of both models as end members in a continuum of possible solutions. Wetting and drying zones that appear in both models are probably more reliable than those that do not.

A comparison of Figures 4 and 5 shows that, in general, the saturation changes from Model 2 are stronger than those from Model 1. The "finger "anomaly observed below the invert of the HD is present in both figures. The dry zones around the wing heaters are also present in both models. Wetting zones above the outer edge of the wing heaters can be observed on the upper left hand comer of the AOD 1 and AOD5 planes; these wetting zones are present in both models. We suggest that these drying and wetting zones are probably real because their presence is insensitive to whether Model 1 and Model 2 are used. Conversely, the dry zone located above the crown of the HD (upper set of images in Figure 3.4-4), is not present in Figure 3.4-5; thus, this dry zone may not be real.

All of the saturation estimates presented are considered to be approximations. The accuracy of the saturation estimates may be limited by one or more of the factors listed in Blair et al, 1998, pages 2-9 and 2-10.

References: Blair, S., T. Buscheck, L. DeLoach, W. Lin and A. Ramirez, 1998, Single Heater Test Final Report, UCRL-ID-131491, Lawrence Livermore National Laboratory, Livermore CA; pages 2-5, 2-6, 2-7, 2-9, 2-10, and Figure 2-4.

Wagner, R., 1998, Drift Scale Test Progress Report No. 1 (BABOOOOOO-01717-570000004, Draft A), August 1998, TRW Environmental Safety Systems Inc., Las Vegas Nevada, pages 7-13 to 7-21.

3-26

Heating day = 406

Date= 1/13/99

Plane dosest to bulkhead, AOD1

Plane intersecting the Heated Drift near the middle, AOD5

3rift

ResistMty Ratio

e rift

0.50 0.80 1.00 1.20

Figure 3.4-1. Tomographs of electrical resistivity ratio for heating day 406. The tomographs show changes relative to the preheating (November 18, 1997) electrical resistivity distribution. A resistivity ratio equal to 1.0 indicates no change; values less than 1.0 indicate that the resistivity is decreasing relative to the baseline. The thick white lines extending from the circle representing the HD represent the location of the wing heater boreholes. The location of the bulkhead and the concrete lined portions of the drift are shown for reference.

3-27

A

I!

Heating day = 461

Date = 3/09/99

Heated [flit

0

D riftU

Resistivity Ratio

Plane intesting the Heated Drift near the middleý, AOD5

.0'

0.60 0.80 1.00 1.20

Figure 3.4-2. Tomographs of electrical resistivity ratio for heating day 461. The tomographs show changes relative to the preheating (November 18, 1997) electrical resistivity distribution. A resistivity ratio equal to 1.0 indicates no change; values less than 1.0 indicate that the resistivity is decreasing relative to the baseline. The thick white lines extending from the circle representing the HD represent the location of the wing heater boreholes. The location of the bulkhead and the concrete lined portions of the drift are shown for reference.

3-28

I

Plane'dosest t -bulkhead, AOD

: •••,.0•• , 11- ... WIE 1"0,• , IV5.50 1W W,. . 00 WRM O M O A -, M,

!I.

Heating day = 406

Date= 1/13/99

0 0Scale 0 S 10 :1.5; 20m

Plane dosest to bulkhead, AOD1

Plane intersecting. the Heated Drift near the middleý AOD5

rift

t

Saturation Ratio

.I I ' I 0.2 0.4 0.6 0.8 I 1.0 1.2

Figure 3.4-3. Tomographs of saturation ratio for heating day 406. The tomographs show changes relative to the preheating (November 18, 1997) saturation distribution. A saturation ratio equal to 1.0 indicates no change; values less than 1.0 indicate that the saturation is decreasing relative to the baseline. These results are calculated assuming Model 2 (the primary conduction pathway is the electrical double layer) (see Blair et al., 1998, Chapter 2, pages 2-5 and 2-6). The location of the bulkhead and the concrete lined portions of the drift are shown for reference.

3-29

Drift

Heating day = 461

Date = 3/09/99

0 S 1 1 20Z~n

0-

Plane dosest to bulkhe'ad,AOD1

trifaJ

Saturation Rati

Piane intersecting the Heated Drift near the m-iddlGe AOD5.

Man]

0.2 0.4 0.6 0.8I . 1. 2

1.0. 1.2

Figure 3.4-4. Tomographs of saturation ratio for heating day 461. The tomographs show changes relative to the preheating (November 18, 1997) saturation distribution. A saturation ratio equal to 1.0 indicates no change. These results are calculated assuming Model 2 (the primary conduction pathway is the electrical double layer) (see Blair et al., 1998, Chapter 2, pages 2-5 and 2-6). The location of the bulkhead and the concrete lined portions of the drift are shown for reference.

3-30

Heating day = 461

Date = 3/09/99

5 10 15S 20 rn

03

Plane dosest to bulkhead, AOD1 rif

Saturation Ratio

I 0.I "0. 0.2 0.4 0.6 0.8 1.0 1.2

Figure 3.4-5. Tomographs of saturation ratio for heating day 461. The tomographs show changes relative to the preheating (November 18, 1997) saturation distribution. A saturation ratio equal to 1.0 indicates no change. These results are calculated assuming Model 1 (the primary conduction pathway is the electrolyte located away from the pore walls), (see Blair et al., 1998, Chapter 2, pages 2-5 and 2-6). The location of the bulkhead and the concrete lined portions of the drift are shown for reference.

3-31

Plane intersecting Heated Drift near 1

middle, AOD5

.the the

Drf bnýe~

3.5 Ground Penetrating Radar

GPR data were acquired in Borehole Numbers 65-68 and 49-51 with no downtime or failure due to high temperature. The antennas were moved smoothly and efficiently within the boreholes and there were no significant problems encountered with the radar antennas over the course of their extended stay at high temperature. The problem in Borehole 48 persisted. Due to a blockage, break, or tight deviation, this borehole was unusable during the most recent data collection visit. The condition of the borehole is currently being examined and when it is judged to be free of any obstructions it will be revisited and used to acquire the radar dataset previously collected between borehole numbers 48 and 49.Processed data continues to show promise. Regions showing sharp velocity increases and decreases (ie., decreases and increases in the dielectric constant of the rock mass) are evident in the majority of the tomograms - suggesting the effect of either temperature or moisture redistribution or both. We have not yet settled on the most desirable method for removing the effect of temperature on radar wave velocity (and hence, the derived dielectric constant). After a lengthy discussion on this topic at the Las Vegas Thermal Test Workshop on April 27, 1999, we have decided to make a first attempt to compensate for the temperature effect by considering only the change in the dielectric constant of the formation water as a function of temperature since the temperature effect resulting from the heated rock matrix is now believed to be quite small. Such values are well established from laboratory measurements made at a sufficiently high frequency. This water vs. -temperature effect will hopefully be accounted for by the time of the next Thermal Test Workshop.

It is clear that the largest regions of high velocity/low dielectric constant (i.e. decreased moisture content) surround the wing heaters and the area closest to the Heated Drift. This, of course, is to be expected. Furthermore, regions of low velocity/high dielectric constant (i.e. increased moisture content) appear to be concentrated directly beneath the Heated Drift and directly adjacent to the aforementioned high velocity/dry rock zone. Additionally, some slight decrease in velocity (i.e. increased moisture content) appears to be evident above the Heated Drift, although the overall decrease is substantially less than the region beneath the drift. A tomogram of radar velocity in the vertical section between Boreholes 65-68 is shown in Figure 3.5-1 (the figure illustrates how the dielectric constant has changed between October 29', 1997 and April 140h, 1999).

Such a distribution of moisture as indicated in Figure 3.5-1 corroborates well with other Drift Scale Test data. Specifically, water production in zones isolated by borehole packers as well as the decreases in the air permeability measurements continue to indicate elevated moisture content in the region beneath the Heated Drift. Some decrease in fracture air permeability was also seen in several of the boreholes above the Heated Drift. Neutron probe data collected in those boreholes closest to the wing heaters continue to indicate large scale drying, as is observed in the radar data. Additionally, the most recent set of ERT data seem to indicate wetting or increased liquid saturation in areas similar to those observed in the radar data. These areas include both the region beneath the Heated Drift and the region directly adjacent to the dry rock zone resulting from the wing heaters, as well as some small increases above the Heated Drift. These drying and

3-32

wetting zones are consistent with the predictions from the thermal-hydrological models. Based on the observations above, it appears likely that the GPR data continue to provide a useful representation of the conditions currently present in the rock mass surrounding the Drift Scale Test.

YUCCA MTN DRIFT SCALE HEATER TEST (GPR RESULTS)

, 10

14 1 5

.0'

0-

Lo

'10

WELLS 10=7 P3 = PRE 04/14/99 = 10/29/97 S 0 -6 -J10 -15 -20 "-25 "| I I ! I I

_0I

-c

IL',ct�

-o

0

In

.- 6 -1 -5 - 0 -2 -So'

-0.01W0.005 0.000 0.0or 0.010 0.01M CHANGE IN VELOCITY (M/NS)

Figure 3.5-1 Change in Saturation as measured by Ground Penetrating Radar.

3-33

-,. F v I

3.6 Neutron data update

Neutron data at the DST, after approximately 480 days of heating, is showing substantial moisture change in only five holes, #50,67,68,79, and 80. They are also referred to as holes ESF-HD-NEU-4, 9 and 10, and ESF-HD-TEMP-1 and 2. We will refer to them as N04, N09, N10, Nll, and N12, respectively. Since last report, at approximately one year of heating, N10 joined the group, with drying first noted at day 457. We now show the data processed using calibration curves measured at a recently completed calibration facility.

The calibration facility consists of 12 cylindrical cells, .9 m diameter by .85m high, filled with granular alumina, hydrated alumina, and sand, mixed to achieve desired densities and- water content. The design water content values were 0, 5, 10, and 15 percent by volume at a density of 2.2 gm/cc and 0 and 10 percent byvolume at a density of 1.8 gm/cc. Half of the cells have a 3-inch hole down the center to model holes NO0 through N1O, the fan shaped arrays drilled from the AOD, and half have a 4-inch hole to model the longitudinal holes NI1, and N12. To simulate the grouted liners emplaced at the DST, we built assemblies of liner material surrounded by grout that would slide snugly into the 3 or 4 inch holes. An additional assembly was built with a full complement of RTD wires to see the effect of the wires in the longitudinal holes.

Calibration curves generated from data taken in the calibration facility on 12/8/98 are shown in Figure 3.6-1 for a density of 2.2 gm/cc. The curve 41in is from the cells with 4" holes. A piece of teflon liner is centralized in the holes but the annulus is empty (no grout). This shows that at zero water content and no grout the probe records a near-zero count rate. The curve 3dst is for the fan shaped arrays of 3" holes drilled from the AOD, NO1-N12. The annuli for these holes are small, about 1/8 inch, but the grout therein still supplies about 200 cps at zero water content in the rock. Curves 4dsta and 4dstb are for the longitudinal holes, respectively with no RTD wires, and with a full complement of RTD wires. The RTD wires displace grout and lower the count rate at a given water content. The effect is fairly constant, with a slight reduction at higher count rates. The number of wires decreases linearly with depth in the hole over the range of temperature measurement. The large amount of grout in these 4-inch holes has a large effect on the count rate, about 800 to 1000 cps at zero water content in the rock. Thus about 60% percent of the measured count rate is due to grout at the nominal rock water content.

Moisture changes in N09 were first seen at day 105. Figure 3.6-2 shows that the dried zone has expanded, and, by day 457 covers approximately 14 meters, from a depth of 15 though 29 meters as measured from the AOD wall. The amount of water in the dried zone drops to nearly zero, with almost all the initial water having been driven away. A later measurement at day 481 shows little change. This hole shows the most and earliest drying because it comes very close to one of the wing heaters, in fact passing between two of them.

3-34

Moisture changes in N04 were not noted until day 300, although there is a slight indication at day 266. This hole is in a similar geometry to hole N09, but does not pass so close to a wing heater. Figure 3.6-3 shows nearly complete drying at day 426 over a 4 m zone from 17 to 23 m, expanding to 5 m by day 455.

Figure 3.6-4 shows clear moisture changes as of day 457 in a small zone at 18 m and further drying to about 40% water loss by day 481. Data from other methods, e.g., radar and permeability, suggest a pervasive increase in moisture deep under the heater drift which is not indicated here. A possible reason is that the permeability data are controlled almost completely by water in fractures, which is a small fraction of the total water seen by the neutron probe.

Hole N11 first showed moisture changes at day 336 as shown in Figure 3.6-5 where we . plot raw count rate instead of water content because it drops below the zero water content

level of about 900 cps on the calibration curve shown earlier. This is apparently the result of the grout drying along with the formation rock, while the calibration curve assumes the water content of the grout remains constant. A substantial part of the water in the grout is chemically bound, and should remain constant. However, If the grout is mixed thinly, as it was for liner emplacement to ensure penetration, more water is available than can be chemically bound and it forms porosity with free water. Such grout is also more permeable, so the free water can escape when heated. We are starting a program to measure the equilibrium water content of grout versus that of rock to sort this out. It is only a problem in the two longitudinal holes which have large annuli. At day 483 much of the water is gone over a 25-meter zone from 25 to 50 meters from the access drift. In addition, a new drying zone is starting at 18 m.

Hole N12 also runs parallel to the heater drift from the access drift, and is the same distance, but on the opposite side of the heater drift from Nll. Although one would think response in these holes would be similar, N12 showed no drying as of day 404. Data taken before the liner was grouted in the hole indicates that the probe should see a count rate of about 600 cps in ungrouted hole. Considering this one can see from the Figure 3.6-6 that the hole is ungrouted at depths more than 52 m, with the amount of grout increasing from 52 m to 48 m, and completely grouted at less than 48 m. Clearly, a substantial amount of the signal in these 4-inch holes is due to grout. The last 4 data sets, day 365,404, 433, and 484, show an accumulation of water in the shallow half of the ungrouted zone. Thin drying zones first show up on day 433, at 20 and 49 m. By day 484 we see drying in 4 zones, 18-21 m, 29-35 m, a small zone at 38 m, and 49-51 m. In addition, a very thin zone at 53 m in the partly grouted zone shows a loss of some of the water accumulated earlier.

3-35

0.2 0.4 0.6 0.8

count rate, kcps

Figure 3.6-1 Calibration curves 12/8/1998.

3-36

1218/98 2.2grncc calibration data

.0

E 0

16

14

12

10

8

6

4

2

0

y -8.9593x2 + 7.4741X - 0.2638 2

R =0.9998

0 4Iin 0 3dst *4dsta

X 4dstb

0 1 1.2 1.4 1.6

0.2

__ 0.18 I 0.16 -

0.14

-_ - -- 105 days E 0,12 1 days

- 230 days - 299 days

0.1 - 327 days U

2 - 352 days "> •-'398

S0.08 - 427 - 457

-'481 0.06

0.04

0.02

-0 0 5 10 15 20 25 30 35 40

depth, meters

Figure 3.6-2 Water volume fraction for borehole N09.

3-37

0.2

0.18

0.16

0.14

-15

1 -65 0~1 2-139

-202

-. 266 01 300

r--329 0C1.C -363

0.08 -399 -4 26

1-455

0.06-

0.04

0.02

0 I I I III

0 5 10 15 20 25 30 35 40

Depth, meters

Figure 3.6-3 Water volume fraction for borehole N04.

3-38

0.2 T

0.18 +

0.16 f

0.14+

0.12+

0.1 +

0.08+

0.06+

F': 0.04-

0.024-

0 I I I

0 5 10 15 20 25

depth, meters

Figure 3.6-4 Water volume fraction for borehole NO 10.

3-39

iC.,.1 I . . i

IIV 1 7. I' F

II.

-70 - 139 -. 201 -265 -327

-- 352 -398

-427

-457 -481

30 35 40

a

1800

1600

1400

1200

1000."- 800

600I-°°

400

200

00 10 20 30

Fi depth, meters

-- Figure 3.6-5 Counts per second for borehole NO1 1.

-78 -99

-141

204 -302

-. 336

-364

-406 -432 -483

40 50

3-40

1600

0 10 20 30 40 50 60

Depth in wall, m

Figure 3.6-6 Counts per second for borehole N012.

3-41

3.7 Isotopic Analyses of Gas and Water Samples

Introduction

Sampling and analyses of the carbon and oxygen isotopic compositions of CO2 continued with another set of gas samples collected from the Hydrology boreholes of the Drift Scale Test (DST) during the beginning of March. These data are being used to determine the sources of increased concentrations of CO2 around the heater drift. In addition, analyses have been made of the hydrogen and oxygen isotope ratios of water and vapor condensate samples collected from the DST. The purpose of these data is to differentiate areas of porewater vaporization and condensation and to assess the degree of dry-out in the system.

Data

C0 2 Analyses: A sixth set of gas samples was collected on March 1-2 of this year. In most of the intervals sampled, the concentrations of CO2 were considerably higher than they were in the previous set of samples (collected during mid-December of 1998). The exceptions to this were intervals close to the Observation Drift and the two intervals that are in the dry-out region (77-3 and 60-2). The intervals close to the Observation Drift probably have a significant level of exchange with the air in the drift. The 8a13C values of the CO2 in those intervals are similar (within 2%o) of the 813C value measured for CO2 from the drift. The low concentrations in 77-3 and 60-2 reflect the diminishing amount of porewater in the rock (which acts as a source of C0 2) after dry out of the rock began (see Figure 3.7-1).

The increased CO2 in the other intervals are a continuation of a trend that started when heating began (Figure 3.7-1; interval 78-3). In the December gas samples, the CO2 concentrations in several of these intervals had not increased significantly or had dropped slightly since the previous sample set was collected in October of 1998. However, it appears that this drop was a temporary effect (possibly due to gas injected into the holes for permeability testing?). In addition, the 513C values of the CO 2 in most of these intervals continued to increase, reflecting a significant level of degassing of porewaters. The exceptions (e.g., interval 78-3) are in areas where the CO2 concentrations are very high (>15,000 ppm). These are areas where it is likely that dissolution of calcite in fractures is contributing to the high CO2 concentrations and it is probable that the measured 813C values are primarily constrained by the 813C values of the calcite rather than by the dissolved inorganic carbon in the porewaters.

1120 Analyses: When they have been available, water samples have been collected from the hydrology boreholes. This set of samples is limited to zones of active condensation with permeability conditions favorable for water to drain into but not out of an interval. To date, water samples have only been collected from intervals 2 and 3 in borehole 60. To augment this data set, vapor condensate samples have been collected from a number of other intervals in the hydrology boreholes. The 5D and 8180 values of the porewater in the rock can then be calculated from the 8D and •180 values of the condensate samples. In order to make these calculations, however, it is necessary to have an accurate estimate of the temperature in the rock. Due to the length of some sampled intervals and

3-42

the large range of temperatures in those intervals, this has made interpreting the data difficult.

Despite these difficulties, several general trends in the isotopic compositions of the waters have become apparent. The initial 8D and 8180 values of the porewaters are about -85%o and -1 1.5%o, respectively (based on analyses of waters spun out of cores taken from the ambient testing boreholes). Water vapor will have 8D and 8180 values less than those of co-existing liquid water (the amount of shift is dependent on the temperature). Therefore, in areas where significant amounts of vapor condensation are adding to the water already in the rock, the isotopic compositions of the water will decrease. The 8180 values of the water samples collected from 60-2 and 60-3 were all lower than -11.5%9 (they ranged between -11.7 and -13.9%o), indicating that significant amounts of steam condensate were present in those intervals. Both of these intervals no longer contain liquid water (60-2 went dry during September - August of 1998 and 60-3 during April of 1999).

The isotopic values of the porewaters calculated from the condensate samples show similar trends in areas where vapor condensation is predicted to be important but water has not accumulated in the boreholes. The 5180 values calculated for porewaters around interval 78-3 are plotted on Figure 3.7-2. As was observed in 60-3, all the values were less than -12%7. Also plotted on Figure 3.7-2 are the calculated porewater 8180 values for interval 77-3. Temperatures in this interval have been above the boiling since August and the 8180 values of the waters have steadily increased since then. This is what would be predicted as the rock dries out. As the lower 8180 steam is separated from the porewaters, the 8180 values of the residual porewaters will increase. Assuming that dry out is essentially occurring at the boiling front (-980C), the last 8180 value calculated for 77-3 (-4.4%7) indicates that approximately 75% of the porewater around that interval has been removed as steam.

Summary

For the most part, the amounts and 813C values of CO2 in the rock around the heater drift continued to increase through March of 1999. The most significant exceptions to this are the intervals where the rock temperatures have passed the boiling point and dry out of the rock is occurring.

The hydrogen and oxygen isotope ratios of the pore waters are also changing significantly. In condensation zones (e.g., intervals 60-3 and 78-3), the 8D and 8•80 values of the porewaters are lower, reflecting input of steam condensates. In the dry out areas (e.g., interval 77-3), the 5D and 8'80 values of the water have increased. Based on the amount of increase in 77-3, it appears that -75% of the porewater in that region has evaporated.

3-43

.D. I1-2.0

-4.0

-6.0

-50.0 -10.0

-14.0

-16.0

-18.0

12/97 2/96 41"8 09 8/93 106/ 12198

Date

Figure 3.7-1. 813C (in %o units relative to VPDB) for CO2 in gas samples collected from interval 3 of boreholes 77 and 78. Also shown in parentheses are the temperatures (°C) and concentrations of CO2 (ppmv).

-4.0

-6.0

-8.0

0

-10.0

-12.0

-14.0 4-12/97 2198 4/98 6198 /98 1eft 12/'98

Date

Figure 3.7-2. 818o values for porewaters around intervals 77-3 and 78-3 calculated from the isotopic compositions of vapor condensate samples collected from the intervals.

3-44

(,9. 23696)

------ ------

"(73-.. 2474M 1189)

78-3 .(4A 14%4)

(39.9 2436)

I p p

2/99

3.8 Water Sampling and Geochemistry Activities

Introduction

During the last quarter of the Drift Scale Test (DST), an extensive revision was made to the aqueous sampling protocols and the corresponding field analysis tests. The revisions addressed previously expressed concerns by geochemists from the thermal test team, and reflected the desire for expanded geochemical data for models' input. Included in the recommendations were proposals for additional field testing equipment, and more detailed specifications on the proper methods of sampling, preserving, and handling of individual samples. The new protocol is a joint effort that incorporates contributions from many of the thermal test team's chemistry and modeling personnel.

In addition to the review of the sampling and testing protocol, water sampling continued to be conducted from wet zones in the Hydrology (HYD) boreholes and the Chemistry (CHE) boreholes (currently under evaluation). The samples were each analyzed for a suite of geochemical parameters. The findings continue to reveal water pooling in certain zones of boreholes directed beneath the heater drift; although some previously wet zones have begun to show drying, consistent with the higher temperatures now recorded. The compositions of these sampled waters have remained fairly stable over time, but the measured pH has been decreasing, a measure consistent with higher CO2 gas concentrations being observed and reported (see Section 3.7).

A summary of the changes recommended in water sampling and field testing from the HYD boreholes will be presented. A summary of water chemistry data will also be reported.

Water Sampling for Chemical Testing

In general, the process of pumping water from a formation at depth and bringing it to the surface produces certain physico-chemical changes in the retrieved sample. These changes may result from several factors: differences in environmental conditions (e.g. temperature, atmosphere, pressure) and exposures to a variety of introduced materials (e.g. pump parts, tubing, filters). To collect a representative sample of the water for analytical testing is therefore a nontrivial task, and the sampling method is often specific to the particular environment and the water chemistry of interest.

In addition to the sampling techniques, the procedures for handling the retrieved sample must be similarly considered. Preventing geochemical changes in the sample, between the time of collection and the time of analysis, requires precautions be taken. Without any special measures, for example, CO2 degassing of the water sample is likely to occur. This loss of CO2 obviously leads to an erroneous determination of the in-situ sample's concentration, but other subsequent measurements are also impacted. These may include pH, alkalinity, and total inorganic carbon; carbonate precipitation may also be induced.

3-45

To ensure a representative chemical analysis is recovered from water samples there are, in general, two types of measures. Preservation techniques may be used at the time of collection, a process meant to halt the chemical changes. Alternately, or additionally, analytical measurements may be made in the field, before the sample has an opportunity to equilibrate. The use of one and/or the other may depend on the chemical parameter to be determined.

HYD boreholes for water sampling

In the Drift Scale Test (DST) of the Exploratory Studies Facility (ESF), several boreholes are instrumented with the expectation of sampling water from different spatial positions in the thermal field. In particular, the HYD holes and the CHE holes are uniquely designed for complimentary water and gas chemistry data collection. Each of the HYD holes has a packer string that inflates to isolate 3-4 open intervals (several meters in length). This instrumentation, previously used during the Single Heater Test (SHT), was intended for air permeability (air K) studies. It was always presumed that the fracture network intersecting the open zones would be pathways for fluids both entering and leaving the cavities, but in one SHT borehole, an unexpected accumulation of water was encountered. In borehole 16, zone 4 (BH16-4), water apparently entered and accumulated in the zone's lower end. When the boreholes were pumped, prior to air K testing, this wet interval yielded several liters of water; this happened on four different occasions. As with the SHT, the HYD boreholes of the DST are again designed and instrumented primarily for the study of air permeability changes. Nevertheless, water-sampling opportunities that present themselves are being exploited to retrieve accumulated water expressly for chemical analyses.

CHE boreholes for water sampling

In addition to the HYD boreholes, CHE boreholes are alternately instrumented with an inflatable liner system designed to allow water as well as gas sampling. The liners were experimental and unproven in a DST-type environment, although they had been effectively used in other geologic settings. Attempts to employ the sampling systems as originally intended exposed several problems. Trouble-shooting efforts were begun, but these attempts proved unsuccessful. As a consequence, an emphasis on an alternatesampling scheme with the emplaced liners was tested and developed. But the methods and the samples are under review and awaiting approval, so the chemistry data and the sample status will be determined and reported at a later time.

Review of the Sampling Protocol

The water sampling details and chemistry reported are being described for the HYD boreholes. The protocol as it is described for sampling in the DST HYD holes is recorded as an R & D entry in a scientific notebook, which is also maintained for documenting the sampling activities. The method describes peristaltic pumping of the open zones through Teflon tubing that is situated in the boreholes between the borehole collar and the lower end of each open zone. After water from a given interval is recovered, the procedure

3-46

dictates splitting the volume into sub-samples for different chemical analyses, and preserving them as necessary. Every sample is given a unique identifier for tracking in accordance with quality control procedures, and a chain-of-custody is initiated.

The sub-samples are designated for specific analyses listed in the protocol. One split is immediately field-tested for temperature and pH, while other splits are filtered and preserved for return to the appropriate analytical service labs. Splits are designated from each sample for the following analyses: major cation and anion chemistry, 0 and H stable isotopes, and Sr and U radiogenic isotopes. The order also reflects the priority of division in cases where collected sample volumes are insufficient for the full analytical suite.

The subject of expanding the aqueous geochemistry parameter list has been a repeated theme among the thermal test group's chemists and modelers. It should be evident from the list reported above, that the carbonate system, for instance, is not well constrained by the current analytical chemistry; but the component is an important parameter in reactive Itransport modeling and corrosion studies. Another very basic parameter not obtained is conductivity, which aids in interpreting the source of the water sampled. Additional parameters (including C isotopes, total dissolved solids (TDS), and colloids) have also been discussed for their particular benefits.

Expanded Geochemistry

The CO2 measure was regarded as the most serious hole in the geochemical data. A basic and representative measurement of the carbonate system is alkalinity, which may be conducted as a field titration. It can be performed immediately after sample collection, minimizing the time of possible exposure and the potential for degassing and concentration changes. It does, however, require proper sampling and handling techniques and therefore a revision to the protocol, as well as acquisition of additional field equipment.

In addition to the carbonate data, measurements of Total Inorganic Carbon (TIC) are desirable in helping to constrain the carbonate system. The measurement requires a carefully collected sample and proper precautions to preserve it until the time of analysis. The C isotopes (14C, and '3C/ 12C) are particularly helpful. The stable isotopes provide information about the age, whereas the radiocarbon aids in identifying the source of the C. The inclusion of sample splits for the carbon isotopes also necessitate a protocol revision for the sample collection process.

The conductivity parameter is viewed as an important and relatively straightforward measure to obtain. Electrical conductivity is an empirical estimate of total dissolved solids in an aqueous sample. The usefulness of EC in the DST samples is as a control on analysis and conservation; therefore, it may be viewed as providing some internal measure of chemical consistency from the analytical data. A multi-parameter field meter often measures EC, pH, TDS, and temperature. The recommendation to acquire a field instrument of this type would increase the field tests with EC and TDS.

3-47

Results of Aqueous Sampling Review

The notebook entry describing aqueous field sampling was distributed for review, and a trip to the ESF for sampling and observation was conducted. As a consequence, several changes were recommended, based on the need to expand the geochemistry (as indicated), but also to improve the methods used in the field to sample. These were prioritized for either immediate or possible future implementation.

The packer system used in the borehole arrays and the limited water accumulations, in general, fixed the sampling options. The method of sampling by peristaltic pumping was regarded as a suitable method in this case. However, the pump-head and the 1/2"-tubing (O.D.) it accommodated were not viewed as optimal. The use of smaller diameter (3/8"O.D.) Teflon tubing and shorter lengths would help maintain fluid-filled lines during sampling. Nevertheless, several other items were given higher priority.

Items that were recommended for immediate attention include the following: the filtration system - to be replaced by an in-line, disposable type filter field test equipment (and corresponding directions): the titrating equipment and reagents; electrical conductivity (EC) measurement capability (a multi-parameter meter (pH, EC, and TDS) with temperature compensation); a thermistor-type thermometer; and supplemental test kits for cations and anions of interest (to check the stability of ion concentrations over time) configuring an in-line, mid-stream sampling port - e.g. a threeport valve for use with gas-tight syringes to mitigate atmospheric exposure of the liquid sample sampling-supplies: stock of pre-cleaned tubing, filters, collection and storage vessels, nitric acid (for preservation); cooler, etc. - to ensure that the potential for crosscontamination is eliminated and to provide the proper preservation and storage capabilities additional sample retrieved when possible for colloid testing (to be assessed by LANL) a reiteration to follow the prescribed sample preservations and to expedite all sample handling destined for analysis

Corresponding to the recommended equipment and protocol changes, a purchase order and detailed rewrite were completed this quarter. The equipment was purchased and is located and available for use in alcove 5 of the-ESF. The rewrite was entered into LLNL's QA document database as a Technical Implementation Procedure (TIP) and may be used by field personnel. An update to the R&D entry in the scientific field notebook to reflect the changes being implemented was also specified.

Aqueous Geochemical Data

The analytical data being reported in Table 3.8-1 are from the HYD borehole samples. Each of the samples was collected under the original protocol entry in the scientific notebook therefore the data do not include the expanded suite of geochemical parameters. The samples are designated in the table by borehole number and zone (BH# - zone#) and are reported with collection dates. It is evident from the data that HYD boreholes 60-3, 59-4 and 186-3 have consistently been "wet" boreholes. The orientations of BH60 and BH186 are declining and angled below the wing heaters and the main heated drift. BH59

3-48

is inclined slightly and angled just above the heaters. BH59 chemical data appear to reflect some contamination of the water from an unidentified source. The trend over time would suggest that water is becoming slightly more dilute as the zone heats up. One possible consideration is that this has been an artifact of the sampling procedure. Warm air pumped from the zone reaches the cooler temperatures of the observation drift and starts condensing in the sampling tube. The overall impact would be a dilution of the actual aqueous sample during pumping. Evaluations of the chemical data are continuing to be made. At the current time, the status of all data reported has been rendered "TBV."

3-49

S3I I d f w s c i

Table. 3.81 The geochemical data from water samples collected in the HYD boreholes. The status of the data is all "TBV."

sample Identification SMF number collection date field pH field Br (ppm)*

Na (mg/L) Si (mg/L) Ca (mg/L) K (mg/L) M mgL Al (mg/L) B (mg/L) S (mg/L) Fe (Mg/L) Li (mg/L)

CI (-mg/-L) HCO3 F (mg/L) Cl (mg/L) Br (mg/L) S04 (mg/L) P04 (mg/L) N02 (mg/L) NO3 (mg/L)

.Thermal testing o03am waters as recepDro bablaed

20. 0 24.0 20.4 17i24t22 6e13

DT6-2 DST 60-3 DST 60-2 DST 60-3 DST 77-3 DST 59-4 DT5-()HYD HYD HYD HYD HYD HYD HYD SPC00527969 SPCO0527977 SPC00527915 SPC00527916 SPC00527917 SPC00541803 SPC00541803 6/1498 6/4/98 81129 8/1 2/98 8/12/98 11/112/98 11/121298 7.5 7.7 6.9 6.8 5.5 6.63 6.63 0.75 0.96 0.385 0.345 probable S~condensation

20.0 24.0 20.4 17 2.4 22.6 135 56 41 52 44 1.48 34 44 20 25 19.9 19 2.09 4,76 450 6.0 4.5 5.4 4.5 1.4 29.5 37.8 2.9 5.7 1.21 4.0 0.21 64.1 83.9 0.12, 0.071 0.017 n.d.< 0.06 0.003 n.d.< 0.06 0.01 n.d. < 0.06 1.2 0.92 1.84 1.1 0.13 4.47 4.13 5.5 9.2 4.5 5.2 1.4 50.7 64.8 0.04 n.d. < 0.02 0.02 0.12 n.d. < 0.02 n.d. < 0.02 n.d. < 0.02 0.07 0.07 0.03 0.04 n.d. < 0.01 0.21 0.20 0.18 0.34 0.11 2.2 0.05 4.02 3.71

1.00 10

0.84 17

0.82 16

0.73 30

n.d. < 0.07 n.d. < 0.07 n.d. < 0.01 n.d. < 0.01

3.00 3.6

0.71 6.14 0.05 4.88 0.25

n.d.<0.04 0.46

0.46 0.6

0.43 5.52 0.21 8.81 0.16

n.d.<0.04 0.6

0.41 2.15 0.03 1.86 1.06

n.d.< 0.4 0.22

0.79 4.34 1,130 1,250 1.13 n.d. < 0.07 226 213

n.d. < 5 n.d. < 0.2 04 n.d. < 3 n.d. < 10

3.12 7.81

3-50

Table 1. Continued. Data is all 'TBV."

sample Identification SMF number

as received lab filtered as received DST 60-3 DST 60-3 (a) DST 186-3

HYD HYD HYD SPCO05418O4 qprnnr.AnA on

IdU [111B (7IdLJ iIIi�f2O

a.s� -- - --

HoY-D3 ta) HYD UD1 6U-3 HYD

________ _ ........ .. . ... ...- o•.uoU. iouo W.00541805 SPC00504396 SPC00504396 SPC00527961 collection date 11/121/98 11/12/98 11/12/98 11/12/98 1/26/99 1/26/99 1/26/99 field pH 6.92 6.92 6.83 6.83 7.4 7.2

Na (mg/L) 10.1 20.3 105 17 19.1 219 25.9 SI (mg/L) 60 54 16 27.2 65 12 49.3 Ca (mg/L) 15.3 13.9 11.5 20.2 5.9 429 2.92 K (mg/L) 8.7 7.8 3.5 3.9 4.1 29.7 5.9 Mg (mg/L) 3.4 3.0 5.1 5.7 1.2 164.0 6.3 Al (mg/L) 0.033 n.d. < 0.06 n.d. < 0,003 n-d. < 0.003 n.d. < 0.06 0.086 n.d. < 0.06 B (mg/L) 1.58 1.41 0.51 0.58 1.75 6.68 0.84 S (mg/L) 11.6 10.5 8.47 9.42 6.4 109 7.9 Fe (mg/L) 0.02 n.d. < 0.02 0.02 0.02 n.d. < 0.02 n.d. < 0.02 0.09 Li (mg/L) 0.04 0.04 0.05 0.05 0.02 0.33 0.05 Sr (mg/L) 0.22 0.20 0.30 0.34 0.09 5.84 0.37 HCO3

41 116 F (mg/L) 0.49 0.5 0.56 0.62 1.27 0.51 1.2 Cl (mg/L) 20 20 19 19 10 1,160 23.30 Br (mg/L) 0.60 0.51 0.67 0.6 0.15 1.51 0.32 S04 (mg/L) 30.6 30.8 26.3 26.2 13.5 240 21 P04 (mg/L) n.d. < 0.2 n.d. < 0.2 n.d. < 0.2 n.d. < 0.2 n.d. < 0.05 n.d. < 0.5 n.d. < 0.1 N02 (mg/L) n.d. <.10 n.d. <.10 n.d. < .10 n.d. <,10 n.d. < .03 n.d. <.3 n.d. < 0.05 N03 (mg/L) 3.38 3.17 7.47 7.27 2.56 11.6 6.73

LDST 59-4 HYD

DST 186-3 HYD

3-51

Table 1. Continued. Data status is "TBV."

3-52

3.9 Acoustic Emission/Microseismic Monitoring:

The acoustic emission monitoring system being operated in support of the Drift Scale Test continues to function without downtime or failure. The monitoring effort continues to record high quality microseismic events with a minimum of false triggers resulting from high frequency electrical noise. In short, the most recent set of modifications and upgrades made to the system have been a success.

The most recent set of microseismic events have been located spatially relative to the Heated Drift. These locations are included and are shown in the attached Figures 3.9-1 and 3.9-2. As can be seen, the events recorded thus far are clustered exclusively above the Heated Drift. There does not yet appear to be any one particular area to which these events are confined. Further locating of events could, however, isolate a region in which a larger degree of activity is occurring.

The presence of the microseismic activity above the Heated Drift should be compared with some of the other results presented at the Thermal Test Workshop, specifically the thermal and power data, the mechanical measurements, and oddly, the neutron logging measurements. Sandy Ballard presented the thermal and power data which seemed to suggest the generation and migration of a steam front above the drift. Any large volumes of steam moving through the fracture systems might result in some slight expansion or "cracking" of the rock matrix and, hence, a microseismic signal. The data presented by Steve Sobolik also suggest some stress relief in the up-going boreholes above the drift. This, too, may be reflected in the acoustic emission data. A plan is currently underway to meet with Steve and discuss the relationship between the microseismic data and the MVPBX data. Finally, the neutron logging results clearly show a zone of wetting and/or water pooling at the end of Borehole Number 80 (located parallel to and above the Heated Drift). This water may in fact be the result of the steam front indicated by Sandy's data and its condensation in the borehole. Again, the rapid upward migration of a large volume of steam might be tied to some of the microseismic events located above the Heated Drift.

Acoustic emission activity will continue to be monitored with the remainder of the recorded events, as well as all new events, being located spatially with respect to the Heated Drift. No other changes to the system or to the approach are currently envisioned.

3-53

I I I I I I

CD

ci".

0

0

0

0

Y>n

Y" (METERS),

II p

Wi:

a

x

a-

K-I :1

S p * S

0

LOCATIONS

30.0

ma

-2.0

-3.0-

0

X 0.-r0S0D .0 4.0 4.o 30.0

Figure 3.9-2 Location of recorded events projected onto an x-z plane.

3-55

N,

3.10 Video Imaging Inside the Heated Drift

No video imaging was completed since Progress Report #2. Additional video imaging is scheduled to be completed in the next few months.

3-56

3.11 Conduction and Convection Through the DST Bulkhead

The loss of heat through the DST bulkhead has been an ongoing concern since the planning and design phases of the DST. Heat loss can be divided into conductive and convective fluxes. Conductive heat loss occurs through the bulkhead's steel construction; whereas the convective heat loss is from water vapor escaping through bulkhead leaks such as the power cables, sensor wiring, doorways, and periphery. It appears the measurement of convective heat fluxes is more difficult than the measurement of conductive heat fluxes. The following discussion provides a chronology of activities associated with the bulkhead's heat loss.

Design Considerations

The bulkhead was designed to perform as a thermal barrier but not as a hydrological barrier. Water vapor was not intended to be trapped. Even though the bulkhead was designed to be a thermal barrier, it was never intended to be perfectly insulated. Consequently, some heat loss was anticipated. This condition is considered acceptable because of the ability to numerically simulate the heat flux and limitations of constructing a thermal bulkhead.

Other design aspects of the DST need to be considered when evaluating the impact of heat loss through the bulkhead such as the existence of an open system in the DST block. An open system is known to exist because of the negligible retardation in barometric pressure between measurements in the local rock mass and the north portal pad. The existence of an open system in the fracture network provides implies numerous pathways, in addition to the bulkhead, for water vapor movement. The DST was designed to overdrive heating in order to expedite the test. This fast heating rate results in additional heat loss through the bulkhead. Also, it was anticipated that numerical simulations/modeling of the DST could accommodate uncertainties, such as bulkhead heat loss, through implementation of suitable boundary conditions and proper sensitivity analyses. Furthermore, the DST design anticipated the need for refinements in the test such as those associated with bulkhead heat loss.

Initial Observations

Shortly after the DST heaters were activated, moisture accumulations on the bulkhead's cool side were observed. Investigations of this phenomenon resulted in an understanding that the moisture was largely condensed water vapor that escaped the bulkhead. The observed moisture, estimated to be 100s of liters, has been a small fraction of the estimated 10 million liters of water mobilized in the test block. Figure 3.11-1 shows graphically another observation stemming from this initial observation which is the inverse relationship of barometric pressure and relative humidity measured in the heated drift. This "barometric pumping" retards the flow of water vapor through the bulkhead, which is a measure of convective heat loss, during high pressure days. Conversely, the flow of water vapor through the bulkhead increases during low pressure days.

These initial observations led to installation of additional thermal and moisture probes along the roof's centerline on the bulkhead's cool side. These instruments facilitate the interpretation of moisture accumulation on the outside of the bulkhead. Also, these initial observations provided insights on

3-57

repository performance including the potential for natural removal of heat and moisture as well as the likelihood of low relative humidity in the heated drift. Refinements

Several refinements in the DST have either occurred or are anticipated. Specifically, baffles have been placed over the ventilation outlets near the bulkhead to reduce the amount of forced convection on the bulkhead. Water vapor leaks in the bulkhead, such as those in the camera door and cable outlets, have been sealed to the extent practical. Sealing is intended to mitigate convective heat loss through the bulkhead. A water collection system was developed to estimate convective heat losses through the bulkhead. Currently, improved methods for measuring conductive and convective heat losses through the bulkhead are being evaluated.

Recent Observations

Conductive heat losses have been measured four occasions as shown in Figure 3.11-2. Results indicate the estimated conductive heat loss through the bulkhead ranges from 5 to 7 kW. Similarly, convective heat losses have been measured from nine different samplings taken from the water collection system. As shown in Table 3.11-1, the convective heat loss through the 1.5 inch-diameter opening in the bulkhead ranges between 0.2 and 0.6 kW. Total convective heat loss is estimated to range from 4 kW to 30 kW. Other observations indicate the presence of a convection cell around the bulkhead and a transient drying trend in the heated drift.

Future Activities

Future activities include ongoing monitoring of the thermal-hydrological behavior in terms of measurements and numerical simulations. This activity includes sensitivity analyses to better determine the impact of heat loss through the bulkhead on the ability to replicate the T-H behavior. Also, existing methods for measuring conductive and convective heat loss through the bulkhead are being re-evaluated to improve accuracy.

Table 3.12-1 Convective Heat Loss from Vapor Removal System.

3-58

Relative Condensate Convective Date HD Air Ternp Humidity Air Pressure Removed Heat Loss (OC) (%) (KPa) (mUhour) (kW)

07/29/1998 133 15.2 90.0 305 0.19 08/0511998 126 8.6 90.4 600 0.38 08/26/1998 135 11.4 89.9 870 0.54 08/31/1,998 135 10.8 90.3 800 0.50 11/10/1998 145 5.1 90.7 600 0.38 02109/1999 159 7.7 90.1 700 0.44 04/19/1999 169 5.0 NA 290 0.19 04/21/1999 169 6.4 NA 700 0.44 04/22(1999 169 6.4 NA 705 0.44

Humidity and Air Pressure in the Heated Drift

0 50 100 150 200 250 300

Time (Days)

- - -..... • 92

S91

_ _90

88:

87

9 I 86 •0.

ft 84

-83

S-1 8 2 350 400 450 500 550

Figure 3.11-1 Inverse relationship of barometric pressure and relative humidity in the heated drift.

3-59

0

4-a

E

4.'

100

90

80

70

60

50

40

30

20

10

0

Figure 3.11-2 Measured convective loss on the bulkhead.

3-60

Location Heat Flux (W/m2)

05-Jan-99 23-Mar-99 26-Apr-99a 26-Apr-99b

1 242 296 277 300 2 344 310 316 338 3 425 594 541 519 4 360 721 699 744 5 592 767 821 823

Est. Total Conductive Heat 5.1 7.0 6.9 7.1

Loss (kW) III

3.12 Interpretive Analysis of the Thermal Hydrological Processes

At the April 27, 1999 Thermal Test Workshop, we presented a comparison of measured data, collected through 15 months of heating, from the Drift Scale Test with simulated results. The spatial and temporal evolution of data pertinent to thermal-hydrologic processes are (1) temperature and (2) moisture distribution in the rock mass. The temperature is, to the first order, controlled by heat conduction. On the other hand, effect of thermal-hydrologic coupling (from fluid movement) on temperature is understood to be of a higher order. The most prominent temperature signature from thermal-hydrologic coupling is that of a heat pipe, i.e., temperature readings persisting at the nominal boiling temperature of - 97`C, indicating the presence of both the liquid and the gas phase. Temperature will only rise above boiling when all mobilized water had boiled away. To track the moisture distribution in the matrix pores, geophysical methods of neutron logging, cross-hole ground penetrating radar and electrical resistivity surveys are being conducted at 1 to 3 months intervals. Air-permeability measurements are also being carried out periodically to track the liquid saturation changes in the fractures.

A comparison of data with simulation up to 15 months leads to conclusions that are similar to those derived from 12 months of heating data. As reported previously, they are: * There is good agreement between temperature measurements and simulations in -1650 temperature sensors installed in 26 boreholes grouped in five arrays emanating from the Heated Drift. The mean error between simulated and measured temperatures has increased from 1.8 0C at 12 months of heating to 3.2 0C at 15 months of heating. * As to the spatial and temporal evolution of moisture distribution, simulations predict a general trend of drying in the vicinity of the Heated Drift and the wing heaters, wetting from the condensation of vapor further away, and increase with time of these drying zones around the heat sources. These are in general agreement with the geophysical and air-permeability measurements. * Local heterogeneity (drainage in fractures) can impact temperature at respective sensor locations.

To obtain an objective measure of goodness of fit between data and simulations, we have computed the Root Mean Square Error (RMSE) and the Mean Error (ME), utilizing the measured and simulated temperature values in all sensors in the 26 boreholes (Table 3.12-1). To avoid bias, we employed a weighting scheme of inverse proportionality to the number of temperature measurements in an interval of-100C. Table 3.12-11 shows the statistical measures for 3-D simulations using an effective continuum approach, for 3, 6, 9 12, and 15 months of heating.

The overprediction of temperature by the thermal-hydrologic model may be attributed to the fact that the model does not yet account for all the heat loss from the test. For example, the model assumes a perfectly insulating bulkhead while measurements show that conductive heat loss through the bulkhead may be several kilowatts. In addition, though the model assumes an open boundary for the Heated Drift wall and bulkhead, giving rise to a convective heat loss of -22kW, it does not account for the effect of strong ventilation in removing additional vapor from the DST.

Comparison between modeled results and measurements is illustrated in Figure 3.12-1, showing the temperature profiles along the eight boreholes (158 -164), at 15 months of heating. The measurements are on the left and the simulations are on the right. Temperature significantly exceeding 100*C are seen in boreholes 160 and 164 up to - 12 m from the collar. These are the two horizontal boreholes parallel to and slightly above the wing heaters. There is a subtle heat-pipe signal at ~ 12 - 13 m in these boreholes,

3-61

Table 3.12-1. Statistical Measure of goodness of fit: the Root Mean Square Error and Mean Error from simulated and measured temperatures of the 26 boreholes (133-134, 137-144, 158-165, 168-169, 170175) in the Drift Scale Test.

Criterion Temp. 3 months 6 months 9 months 12 months 15 months range

ME All 0.85 0.83 1.25 1.82 3.2* RMSE All 4.31 5.81 7.06 8.14 11.63**

ME < 97*C 0.90 0.75 1.25 1.85 3.4 RMSE < 970C 4.33 3.93 3.92 4.42 10.22

(1261 sensors)

ME >970C -.87 1.43 1.26 1.71 2.28 RMSE >970C 4.15 8.44 9.51 10.46 5.56

(378 sensors)

*Positive Mean Errors (simulation minus measured) indicate error in energy balance ** Large Root Mean Square Errors indicate effect of heterogeneity

indicating that around the tip of the wing heater a two-phase vaporization zone still persists. Not only that, the existence of a two-phase zone from vaporization due to heat (from the Heated Drift as well as the wing heaters) and condensation in cooler region is seen in the heat-pipe signature in boreholes 161, 163 and 165 at - 2 - 4 m from the collar. All the above signatures of thermal hydrological coupling in the temperature data are reproduced in the simulations.

Drying and wetting in the DST block is illustrated in Figure 3.12-2, where we show the simulated liquid saturation in the matrix (left) and the fractures (right) at 15 months of heating. Liquid saturation is expressed in a color palette ranging from 0.2 (red) to 1.0 (blue) for the matrix, and from 0.05 (red) to 0.5 (blue) for the fractures. Neutron logging and cross-hole radar surveys were carried out in boreholes 64 through 68. They indicate zones of drying consistent with that simulated for the matrix liquid saturation. Similarly, air permeability tests were conductedin boreholes 74 through 78 and the zones of decreased air-permeability coincide with the simulated zones of increased fracture liquid saturation.

Because of parameter uncertainties, two different hydrological property sets were included in the pretest predictive model (Birkholzer and Tsang, 1997). The early time DST data seem to favor the hydrological property set that corresponds to a lower percolation flux. As a result, we have taken that property sets as the reference set for our simulations. We continue to study how the DST data can serve to constrain the hydrological and thermal properties and have begun to investigate the sensitivity of DST results to thermal conductivity values of the rock mass. While our model uses the heat conductivity values of 1.67 WIm OK and 2.0 W/m *K respectively for "dry" and "wet" rock mass (these values are specific to rock cores from the Single Heater Test block), Total System Performance Assessment has used 1.56 W/m OK and 2.33 W/m OK for the Topopah Spring middle nonlithophysal welded tuff.

3-62

Preliminary results based on simulation studies, show that incorporating these new thermal conductivity values can reduce the mean error between simulated and measured temperature. We shall continue our investigation to utilize DST data to help reduce parameter and conceptual model uncertainties.

Reference: Birkholzer, I.T. and Y.W. Tsang, 1997. Pretest analysis of the thermal-hydrological conditions of the Drift Scale Test at Yucca Mountain, LBNL-41044, Lawrence Berkeley National Laboratory, Berkeley, California.

200.0 200.0

190.0 190.0 0 153 JO) 1 0 159 (45') 0.0----- 153 (0') 10.0 --- 59 (45') 170.0 ,V 160 (90') 170.0-4- 160 p0 161 (135)

16 (135') 16. 11 162 (080'> 160.0 .~ . 162 1 0' 62 1(2 (100 150.0 40 163 (-1315') 150.0---- 163 (.1O") 1100 64 (-45') C)000--- 164 (-100')

164 (.90') ,"• 164 (.90')

U ' 1 40 .0 1 6- ( 4 5 U 4. 0 • 16 5 (-4 5 ') t-130.0 1030.0 1.

620.0 ~~oooo 12100

0.0.0 200.0 8 0.0 70.0 Irv 0.0 70.0 70.0

50.0 50.0

40.0 40.0 30.0

30.0 20.0 0.00

0.0 50 0.0 15.0 20.0 2. 0.0 5'0 10.0 15.0 20.0 Distlance from Borehole Collar (in) Distance from Borehole Collar (m)

Figure 3.12-1. Temperature profiles in boreholes 158 -164 at 15 months of heating. Measurements (left), simulations (right).

10.

0.70

04

-10.0 ______________________________ -10.0

1-00 --. O .O0 V0.10 -20.00.

x [Ir] x [m] Matrixulquld Satumraton Eracture liquid Saturation

Figure 3.12-2. Simulated liquid saturation at 15 months of heating in the matrix pores (left) in a xz vertical section containing boreholes 64 through 68, and in the fractures (right) in a xz vertical section containing boreholes 74 through 78.

3-63

10.

0,0.j

3.13 Thermo-hydro-chemical Analysis Model

Introduction

This section describes modeling results of thermal-hydrological-chemical processes for the Drift Scale Test (DST). The modeling studies allow for a quantitative assessment of chemical processes coupled to the thermohydrological driving forces. Comparisons are made to gas phase CO2 concentrations collected approximately 15 months after initiation of the test (section 3.7) and the chemistry of waters collected 12 and 14 months into the heating cycle (Section 3.8).

Water Chemistry

The average matrix pore water composition from alcove 5 is shown in Table 3.13-1 (Section 3.8). Also shown are chemical compositions of four waters collected from hydrology boreholes 60 and 186 on December 11, 1998 and January 26, 1999 (Section 3.8). The latter samples were obtained from zones that were hotter than the temperatures given for the samples, because water temperatures were measured after collection and had cooled substantially. Both intervals are located in the zone below the wing heaters. Borehole interval 186-3 is lower than 60-3 and waters were probably cooler from this interval.

Some clear trends in water chemistry in both intervals over time are increases in pH and SiO2 (aq) concentration, and a drop in Ca. A similar trend for pH and SiO2 (aq) exists between the boreholes, where the hotter interval (60-3) has a higher pH and SiO2 (aq) concentration than the 186-3 interval at each time. The concentration of HCO3 is also lower in 60-3 relative to 186-3 as expected from the * higher temperature of 60-3. Other relationships are less obvious.

Some of the processes that must be deciphered from these data include mixing of pure condensate water with fracture pore waters, equilibration of condensate with matrix pore waters via molecular diffusion, reaction of condensate waters with fracture-lining minerals, and mineral precipitation due to reaction, boiling, temperature changes, or pH changes. The high silica concentration, relative to chloride and the initial pore water silica concentration is consistent with dissolution of a silicate phase, rather than increased concentration by boiling. However, concentrations of K, Mg, and Na are higher than what would be expected by dilution of original pore water (as evidenced by the low chloride concentrations). Therefore, the silicate phases that dissolved must have been some combination of cristobalite, opal, feldspar, clays or zeolites, rather than just a simple silica mineral containing just SiO2.

THC Model Simulation Parameters

The simulation described here used the dual permeability formulation as applied to reactive transport (Sonnenthal et al., 1998a) and thermohydrological processes (Birkholzer and Tsang, 1998). Hydrological parameters are those presented in Birkholzer and Tsang (1998) and were calibrated for Borehole SD-9 at an infiltration rate of 0.36 mm/year. Initial conditions were based on a steady-state simulation using these parameters. Based on measurements of fracture porosity in the DST (Freifeld and Tsang, 1998) the fracture porosity was increased by a factor of ten over the base case value to approximately 0.0026. THC calculations were performed using TOUGHREACT v2.1 (Xu and Pruess, 1998; Spycher et al., 1999). The simulation was run with a pretest ventilation period of nine months, followed by 2 years of heating at full power. Results are presented after 12 and 15 months of heating, for which measured chemical data were available.

3-64

Table 3.13-1. Measured concentrations in TSw pore water from Alcove 5 (used as initial fracture and matrix pore water) and chemistry of water taken from hydrology boreholes.

Parameter

emperatui

H

a SiO2 (a ) Ca K Ag Al

HC03 '

7-..Unit

S

C

mg/l

mg/l 'mg/1

mg/if

Pore Water

(1)25

8.32

61 71 101 8.0 17

lxl0"',

20 n.a 41

60-3 (12/11/98)

26.5-49.6

60-3 (1/26/99)

51.7

186-3 (12/11/98)

34.3-34.8 unknown

6.92 7.4 6.83 7.2

20. 115-.5 13.9 7.8 3

n.d. (< 0.06)

19.1 139 5.9 4.1 1.2

n.d. (< 0.06)

U1

S04

C02 (gas) (4)

mal117 20 in 1m~l 116 30 R 1

-y-j 870

17 58.2 20.2

3.9 5.7

n.d. (< 0.06)

n.a.

n1a

26.2

n.a.n.a. n.a.

(1) Average of porewater analyses ESF-HD-PERM- 1 (30.1 '30.5') and ESF-HD-PERM-2 (34.8"35. 1'). (2) Calculated by assuming equilibration with Ca-smectite at 25 C. (3) Total aqueous carbonate as HCO,. calculated from charge balance. (4) Calculated at equilibrium with the solution at 25 *C.

The chemical processes considered are the outgassing, transport, and redissolution of CO2 and the precipitation/dissolution of calcite, silica phases and gypsum, and transport of aqueous species. Minerals, aqueous species, and gas species used in the simulation are given in Table 3.13-2. All phases except gypsum, which is considered to react under equilibrium conditions, undergo kinetic dissolution and/or precipitation. Feldspars, clays, and zeolites are not included in this simulation as their reaction rates are extremely slow and apparently have little effect over short time scales on the modification of pH, PCO2, porosity and permeability. Simulations involving these phases were presented in Sonnenthal et al. 1998) and Spycher et al. (1998). Thermodynamic and kinetic input data are summarized in Sonnenthal et al. (1998) and Spycher et al. (1998). The initial water composition in fractures and matrix is taken to be the average measured pore water composition given in Table 3.13-1. The initial gas CO, concentration was calculated based on equilibration with this water at 25 °C.

3-65

186-3 (1/26/99)

7 25.9105.5 2.92 5.9 6.3

n.d. (<

116 23.3 21

n.a.

mz/1 116 1•

200 41hALf

rn eA 117 20

Table 3.13-2. Mineral phases, aqueous and gaseous species used in DST model simulation.

THC Model Results and Comparison to Measured Gas and Water Chemistry

Simulation results for PCO2 after 15 months of heating are contoured in Figure 3.13-1 with comparison to measured PCO2 (Section 3.7). Regions of increased PCO2 are similar in location and magnitude to those measured, except the inner edge of high PCO, begins slightly further out from the heaters in the model simulations. This is most likely due to the higher temperatures predicted from the 2-D simulation compared to measured temperatures and those predicted from the 3-D models. The outer zones of increased PCO, are well-captured along the transect of borehole 74, where there is a progressive increase out from the observation drift, with a decline in the last interval (74-4). Around the observation -- drift there appears to be much more inward progression (toward the heaters) of ambient CO, into the system compared to the model predictions, as well as more buffering of drift air and the surrounding dryout region. Effects of barometric pumping and pressure changes due to ventilation may be responsible for these effects that are not considered in the model simulations.

The pH of waters in fractures is shown in Figure 3.13-2. A declining pH from the initial value of 8.3 to around 7 is seen throughout the condensation and drainage zones, with increases toward the boiling front as CO2 is degassed from the system.

Contoured distributions of volume percent changes in fracture calcite are shown in Figure 3.13-3. Dissolution of fracture calcite is well developed in condensation/drainage zones below the wing heaters with less dissolution above the heater drift. Significant precipitation is localized in the dryout zone.

Fracture silica volume percent change (sum of changes in cristobalite, quartz, and amorphous silica) is shown in Figure 3.13-4. In contrast to calcite that shows increased dissolution below the wing heaters, dissolution of fracture silica is greater above and to the sides of the wing heaters. Effects of drainage are not important for silica, because solubility decreases with declining temperature, whereas calcite solubility increases and therefore hot waters draining into cooler regions would tend to dissolve calcite and precipitate silica, if both are are saturated at the higher temperature.

Gypsum is not considered-a primary mineral in the simulations, as it has not been observed as a fracturelining mineral. However, during boiling and subsequent increased concentrations of sulfate, it precipitates as a late-stage phase just prior to dryout. Its distribution in Figure 3.13-5 clearly shows a pervasive zone of precipitation restricted to the dryout zone (above approximately 96 °C).

3-66

Minerals Primary Aqueous Gaseous Species Species

Cristobalite-t SiO (aq) 0O2

Quartz Ca -0 orphous a

ilica Palcite I' Gypsum CO'

,2_

Figure 3.13-6 shows the total change in fracture porosity (sum of calcite, cristobalite, quartz, amorphous silica, and gypsum changes). Increases in porosity (positive values) are strong due to the combined effects of cristobalite, quartz, and calcite dissolution giving a more uniform band around the wing heaters, and less so above the heater drift. This porosity reduction is due predominantly to calcite precipitation.

A series of fracture water compositions are shown in Tables 3.13-3 and 3.13-4 for a short vertical section taken from the condensation and drainage zones near intervals 60-3 and 60-4 (see Figure 3.132). The section was shifted slightly downward because of the overpredicted temperatures in the 2-D model, and because water analyses at 14 months are compared to a fifteen month heating period. Basically, waters collected in the boreholes correspond to some unknown volume of draining waters and the model results are fracture waters from an individual grid block. Therefore, the general trends in composition as a function of distance and temperature must be compared and not an exact location.

As in the measured data (Table 3.13-1) the model results show the effect of increasing pH and aqueous silica with time and inward toward the heaters in the high temperature areas (greater than 60 °C) near the 60-3 and 186-3 borehole intervals. Concentrations of HCO3 also decrease toward the higher temperature zone and are in the range observed. Chloride and sulfate show similar ranges to those measured, but in the model results the ratios are constant, because there is no dissolution or precipitation of a sulfate phase in the drainage and condensation areas. There is some variation in the C1/SO4 ratio in the 60-3 and 186-3 waters, possibly due to variability in pore water compositions or the presence of trace sulfate phases in dry regions of the fractures prior to the test. Calcium in the model results tends to be higher than the measured values, especially compared to the samples collected on 1/26/99. This may be the result of greater precipitation of calcite in high temperature waters draining into boreholes after the initial dissolution by interaction of early low pH condensate with fracture calcite. However, it could also signify the precipitation of another calcium-bearing mineral.

Table 3.13-3. Fracture water compositions, pH, liquid saturation, and temperature from simulation (dstjthc9) after 12 months of heating. Data are from a short vertical section near the Borehole 60-3 and 186-3 intervals (illustrated on Figure 3.13-2).

X (in) Z (im) S-LIQ T (C) PH Ca Na Cl1 Si02 HCQ3 S04 -7.72 -4.57 0.26 96.83 8.1110.3 3.06 5.84 69.46 15.41 5.79

4 -7.88 -5.82 0.38 86.867.28 0.2 4.16 7.95 53.93 59.16 7.88

7 -8.21 -7.57 0.4 63.29 7.07 38.6 10.3 19.7 51.28 114.59 19.62 1 7 9 -8.05 -9.82 0.43 44.45 7.12 51.1 16.5 31.5 51.94 144.79 31.24

1 1 -2 -8 -12 0.44 34.41 7.2360.2 23.0 44.0 53.641161.02 436

T8 f6 17 1 '

3-67

Table 3.13-4. Fracture water compositions, pH, liquid saturation, and temperature from simulation (dstjthc9) after 15 months of heating. Data are from a short vertical section near the Borehole 60-3 and 186-3 intervals (illustrated on Figure 3.13-2).

X (m) Z (m) S-LIQ T (C) pH Ca Na Cl Si02 HCO3 S04 -7.72 -4.57 0.13 97.79 8.16 11.14 3.96 7.56 85.44 12.95 7.49 -7.88 -5.82 0.29 96.09 8.06 10.55 2.93 5.59 72.4 17.43 5.54 -8.21 -7.57 0.36 76.13 7.19 31.24 8.78 16.7 56.9 88.96 16.61

-8.05 -9.82 0.39 53.6 7.1 49.3 17.0 2.5 54.05 137.77 32.31 OT9

-8 -12 0.41 40.5 7.16 61.68 25.01r7.7 56.27 164.13 47.341 61

Conclusions

Model results for aqueous and gas phase chemistry seem to reproduce the major trends as deduced from gas and water sample measurements. Measured water compositions from hydrology borehole intervals 60-3 and 186-3 show significant effects of silicate mineral and calcite dissolution, dilution of original fracture and/or matrix pore waters with condensate waters, and interaction with migrating CO2 in the vapor phase. Greater differences in the model results for CO2 near the observation drift and the heater drift indicate that updates of the model should consider greater exchange with drift air.

References Birkholzer, J. T. and Tsang, Y. W. 1998. "Interpretive Analysis of the Thermo-Hydrological Processes of the Drift Scale Test." In: Drift Scale Test Progress Report, Lawrence Berkeley National Laboratory, Chapter 6. Yucca Mountain Project Level 4 Milestone SP2930M4. Berkeley, California: Lawrence Berkeley National Laboratory. MOL. 19980825.0268.

Freifeld, B. and Tsang, Y.W. 1998. "Active Hydrogeological Testing". In: Second Quarter TDIF Submission for the Drift Scale Test (Hydrological, Radar, Microseismic), Chapter 2. Yucca Mountain Project 4 Milestone Report SP2790M4. Berkeley, California: Lawrence Berkeley National Laboratory. MOL. 19980812.0240

Sonnenthal, E.; Spycher, N.; Apps, J.; and Simmons, A. 1998. Thermo-Hydro-Chemical Predictive Analysis for the Drift-Scale Heater Test. Yucca Mountain Project Level 4 Milestone SPY289M4. Berkeley, California: Lawrence Berkeley National Laboratory. MOL.19980812.0268.

Spycher, N.; Sonnenthal, E.L.; and Apps, J. 1998. "Interpretive Analysis of the Thermo-HydrologicalChemical Aspects of the Single Heater Test." In: Tsang, Y.W.; Apps, J.; Birkholzer, J.T.; Freifeld, B.; Hu, M.Q.; Peterson, J.; Sonnenthal, E.; and Spycher, N. Single Heater Test Final TDIF Submittal and Final Report, Chapter 4. Yucca Mountain Project, Level 4 Milestone SPI 190M4 and SPY147M4. Berkeley, California: Lawrence Berkeley National Laboratory. MOL. 19980921.0103.

Spycher, N., Sonnenthal, E.L., Alders, C.F., and Xu, T., 1999. Software qualification of TOUGHREACT v2.1, submitted.

3-68

Xu, T. and Pruess, K. 1998. Coupled Modeling of Non-Isothermal Multi-Phase Flow, Solute Transport and Reactive Chemistry in Porous and Fractured Media: 1. Model Development and Validation. LBNL42050. Berkeley, California: Lawrence Berkeley National Laboratory.

3-69

FRACTURE LOG PCO2: 15 MONTHS

25

20

15

10

5

0

-5

-10

-15

-20

-25-30 -20 -10 0 10 20 30

[meters]

-4.4 -3.8 -3.2 -2.6

LOG PCO2 [bars]

Measured CO 2 On March 1, 1999 (Conrad, 1999) I I I

25i

a)

20

15

10

5

0

-5

-10

-15

-20-

-2.36 -2.96 -

-2.52,74

" 75

-76

Drift)-3.05

-1.36

78

Log PCO2 (bars)-, -., "' I I

-30 -20 -10 0 10 20 30 [meters]

Figure 3.13-1. Modeled PCO2 distribution after 15 months of heating (top). Measured PCO2 in gas (originally as CO2 concentration) taken from hydrology boreholes (Conrad, 1999).

3-70

E

-5.6 -5.0 -2.0 -1.6

m l I II• .tllf.•

FRACTURE pH: 15 MONTHS

I1I I

-30 -20 -10 0 10 20 30

[meters]

6.9 7.2 7.5 7.8 8.1 8.4 8.7 9.0 92 pH

Figure 3.13-2. Fracture water pH distribution after 15 months of heating with temperature contours overlain. Locations of hydrology borehole intervals 60-3 and 186-3 where water samples were collected are shown in white. The blue line and 'points represent a short vertical section through the condensation and drainage zones close to intervals where water was collected.

3-71

25

20

15

10

5

0

-52

-10-"

-20

-25-

Porosity Change [%]: 18 MONTHS

25 • _

20 -

15

1,10D-0 -0 01 0 3

15 - i

Cacie hage(%

E

-25 -30 -20 -10 0 10 20 30

[meters]

-0.006 -0.001 0,004 0.009 0,014 0,019 0.023

Calcite Change (%70)

Figure 3.13-3. Fracture calcite volume percent change after 15 months of heating with temperature contours overlain.

3-72

Porosity Change [%]: 18 MONTHS

25

20

15

10

5

0

-5

-10

-15

-20

-25--30 -20 -10 0 10 20 30

[meters]

-0.018 -0.014[mtes

-0.010 -0.006 -0.002 0.001Silica Change (%)

Figure 3.13-4. Fracture silica (sum of cristobalite, quartz, and amorphous silica) volume percent change after 15 months of heating with temperature contours overlain.

3-73

(1) L. ':1)

E

Gypsum Change [%]: 15 MONTHS

I I I I "1

-30 -20 -10 0 10 20 30[meters]

0.0000 0.006 0.(V�1 I�>J LI �4N�)12 0.018 0.024 0.032

Gypsum precipitation (%)

Figure 3.13-5. Fracture gypsum volume percent precipitation after 15 months of heating with temperature contours overlain.

3-74

25

20

15

10

5

0

-5E

-10

-15

-20

-25

Porosity Change [%]: 18 MONTHS

25

20

15

10-

5.

0

"-5

-10

-15

-20

-25-30 -20 -10 0 10 20 30

[meters] " •. ..•1. t i I 4 , . .. ýI• , ! . .. ` j " ; " !

-0.05 -0.04 -0.03 -0.02 -0.01 0.00 0.01 0.02

Porosity change (%)

Figure 3.13-6. Fracture porosity change after 15 months of heating with temperature contours overlain.

3-75

E