. . . I OFT ORNL P 2006 • . 1. . .; . :.; 4.-. F . TO 125 LA LE MICROCOPY RESOLUTION TEST CHART NATIONAL BUREAU OF STANDARDS -1963 9 . ORNAR-2006 CONF-660208-10 MASTER 198 16 1966 PROJECT SALT VAULT: EFFECTS OF TEMPERATURE AND RADIATION ON PLASTIC FLOW AND MINE STABILITY* R. L. Bradshaw, T. F. Lomenick, W. C. McClain, W. J. Boegly, Jr., F. M. Empson, and F. L. Parker Health Physics Division Oak Ridge National Laboratory Oak Ridge, Tennessee RELEASED FOR ANNOUNCEMENT IN NUCLEAR SCIENCE ABSTRACTS : For publication in Proceedings of International Symposium on the : Solidification and Long-Term Storage of Highly Radioactive Wastes Richland, Washington February 14-18, 1966 LEGAL NOTICE This report was prepared as an account of Government sponsored work. Neither the United Statos, por the Commission, nor any person acting on behalf of the Commission: A. Makes any warranty or reprouatation, oxproused or implied, with respect to the accu- racy, completeness, or unolainess of the Information contained in this report, or that the wo of any information, apparatus, method, or procon dieclound in the report may not infringe prinatoly owned righto; or B. Ammos may tahities with respect to the une of, or for dumnego roruiting from the un oi day taforantion, appuntu, method, or proceu daclound in this report. As und in the above, "person acting on behall of the Commission" includes any on- ployee or contractor of the Commission, or ouaploys of ench contractor, to the extent that such employs or contractor of the Commission, or employme of such contractor propera, dorminatos, or provides access to, nay Information purrupt to his employment or contract wità the Commission, or his employment with such contractor. Research sponsored by the U. S. Atomic Energy Commission under contract with the Union Carbide Corporation. Ni NA .. L . 1 9 . PROJECT SALT VAULT: EFFECTS OF TEMPERATURE AND RADIATION ON PLASTIC FLOW AND MINE STABILITY, R. L. Bradshaw, T. F. Lomenick, W. C. McClain, W. J. Boegly, Jr., F. M. Empson, and F. L. Parker Health Physics Division Oak Ridge National Laboratory Oak Ridge, Tennessee ABSTRACT In mines in flat-lying salt deposits, heaving floors and sagging roofs can prove hazardous. The effects of tem- perature and load stress (in roof-supporting pillars) on such stability problems have been studied by means of model pillars. Also, we have investigated the effects of radia- tion (doses up to 5 x 100 R) on small test specimens. Oper- ation of the radioactive and electrical arrays in Project Salt Vault has not revealed any significant effect of radia- tion on the salt flow or stability (as of January 1, 1966). However, thermal stresses from the heat in the floor have been found to be transmitted to the roof, resulting in a fivefold increase in the rate at which the roof is sagging. No problems due to the increased flow are anticipated, either in the demonstration or in a properly designed actual disposal facility, but the need for careful study of such effects is underscored. 1. SALT MOVEMENT IN MINES In general, when an opening is created in a salt mine, there is a rapid flow of salt into the opening. The rate at which the opening closes is dependent upon such things as the depth of the mine, the percentage of the area which is extracted the extent of the area which has already been mined, the shape and height of the pillars, the temperature of the salt, and the time since creation of the opening. afor publication in Proceedings of International Symposium on the Solidufi. cation and Long-Term Storage of Highly Radioactive Wastes, Richland, Washington, February 14-18, 1966. BResearch sponsored by U. S. Atomic Energy Commission under contract with Union Carbide Corporation. . . . . . . .-*** Asli Roughly speaking, the closure rate is inversely proportional to tine. That is, with ench doubling of the time since the opening was created, the rate of closure is halved. For example, if the floor-to-ceiling closure rate were 0.04 in./day 10 days after mining of the area, it would fall to about 0.02 in./day by the end of 20 days, to 0.01 in./day by the end of 40 days, etc. In flat-lying (beided) deposits where shale is interbedded with the salt, the flow of salt frequently causes separation of layers of salt from the shale layers with resultant floor heaves and ceiling sags. If too much salt is ex- tracted, or the supporting pillars are not properly sized, or if the thickness of the layers of salt in the floor and in the roof (down to and up to the first shale layers) is not great enough, the floor heaves may be large enough to inter- fere with the movement of machinery, and hazardous roof falls may occur. Even in mines where salt flow does not cause trouble during the mining operation, roof falls and floor heaves may occur after an extended i Figure 1 shows an example of a floor heave in an old area of the Lyons, Kansas, mine. Figure 2 shows an area where a section of ceiling has sagged and fallen, leaving a sagging lip around the edge of the room. 2. MODEL STUDIES ON EFFECTS OF TEMPERATURE AND STRESS Sufficient empirical data on existing mines are available to enable one to predict the stability of mines at ambient temperature, but no data or experience are available which are directly applicable at elevated temperatures. Recent work by Serata and Oberta at ambient temperatures has shown that flow in rock- salt mines may be approximated by testing scale-model specimens uniaxially by providing proper horizontal restraint over the floor and roof portions of the model. which We have run tests on scale-model pillars at temperatures up to 200°C and stresses up to 10,000 psi. To simulate pillar, roof, and floor conditions wh would exist in mined cavities in rock salt, sample specimens are fabricated to represent scale models of salt pillars and their surrounding rooms. The test specimens are cylindrical in shape, with a portion of the center ground out to form the pillar and surrounding rooms (Fig. 3). By attaching steel rings around the ends of the samples, confining pressure is applied to the roof and floor portions of the models when they are loaded. The following empirical equations have been fitted to the results of pillar- model tests for times from 10 hr on. ė = 3.2 x 10-36 20.963.2 +-0.65 € = 9.2 x 10-36 p10.9 23.2.40.35 + A where ė strain rate (vertical convergence, uin. in--hr--), € = cumulative deformation (uin. in-2), T = absolute temperature (°K), o = average pillar stress (psi), t = time (hr), and A = constant adjusted to fit experimentai curve at 10 hr. :: F . 10 : " 27 U > - 30% more DA ! ) 2 . Fig. 1. Example of Floor Heave, in Older Area of Lyons Mino (ORNL Photo-66925) r . L' . .. NO Su 4 Liri. c C · ...... . . . " - . . :..0. . : . Do 27 a . a Y és A . VA Fig. 2. Example of Ceiling Slab Which Has Sagged and Fallen. (ORNL Photo-67010) - - - - ORNL-DWG 65-1543 : Fig. 3. Salt Pillar-Model Detail. . . : I!! *1*** ... i i 1.. .... ti . t . WIL . 11 ... to Madh, - - - - PILLAR ROOF FLOOR STEEL RING Billetter 1111111!!! . . . ! ., - - + - N IN . . T 4 . . 'Extrapolation of the rate equation produces predicted creep-closure rates which are in reasonable agreement with rates which we have measured in the Kansas mines since 1959 ref. 4). The relationships between vertical and horizontal closure rates measured in the models have also been corroborated by measurements in the mines. 11 Figure 4 shows a prepared pillar-model before testing. In a similar modei, 1/8-in.-diam horizontal holes were drilled at the top and bottoni of the pillar and filled with colored salt before the sample was deformed about 60% (that is, the floor-to-ceiling dimension was reduced from l in. to about 0.4 in.). After deformation, the model was sawed in half vertically along the plane containing the holes. The result is shown in Fig. 5. It is evident that the salt-filled holes are no longer horizontal, being much closer together at the sides of the pillar. While the room heignt decreased about 60%, the center of the pillar shortened only about 20%. This shows that some of the salt from the regions above and below the pillar has flowed out into the floor and roof regions of the cavity. Strain gages attached to the steel restraining rings in a number of the tests show that about 50% of the axial load stress is transmitted to the rings. The model tests have thus shown, in a qualitative way, wliy floor heaves and roof sags take place. Cavity closures for models axially loaded to 4000 lb/in. and 6000 lb/in.' at temperatures of 22.5°, 60°, and 100°C are shown in Fig. 6 as functions of time. It is clear that, elevated temperature has a marked effect on the defor- mation rates. Note also that the deformational behavior of the pillar loaded to 4000 psi at & temperature of 60°C is approximately the same as the behavior of the sample loaded to 6000 psi at room temperature. In many cases salt is mined to a shale layer (sometimes called "partings") since they form planes of weakness at which the salt will readily separate. The presence of these shale partings can have a significant effect on the rate of cavity deformation, depending on whether the shale is at the top and bottom of the pillar or in the center of it. This is shown in Fig. 7, where deformation curves are presented for model pillars that have been deformed without shale partings, with simulated partings (greased teflon sheets) at the top and bottom of the pillar, and with a simulated parting in the center of the pillar. The actual effects to be expected in a mine would lie somewhere between the two ex- tremes shown in Fig. 7. 3. LABORATORY TESTS ON EFFECTS OF RADIATION ON PHYSICAL PROPERTIES Uniaxial compression testing has been carried out at ORNL to evaluate the effects of radiation on the physical properties of rock salt from the Kansas formation (bedded) and from the dome deposit at Grand Saline, Texas." Salt from s exhibited some minor variations in behavior, but only the Kan- sas salt will be discussed herein. Briefly, the results were as follow: 1. The compressive strength of rock salt exposed to 5 x 10° R is about 10 to 20% less than the strength of unirradiated rock salt. On the basis of sta- tistical evaluation, the observed differences are probably real. 2. The modulus of elasticity is greater aiter exposures of 5 x 10° R than for unirradiated specimens. 3. The irradiated salt is more resistant to creep strain. II e SO . SA tj. f mm LABORATORY - . . H 1. I UN . ? Sheeran + . 1 1 :-** . . . 1.. . . . im . . CII M.4. Pillar-Model Before Testing. (ORNL Photo-67559) .. ..':, : * ento i 10 i 1.1.. . 2.. .3 AV. Fig. 5. Section Through Pillar-Model After 50% Deformation. . . . . . . . ---m.ocn. - - - NR V . . ORNL-DWG 65-4542 6000 psi 100 °C VERTICAL SHORTENING OF PILLARS (in.) 4000 psi 100 °C 6000 psi 60 °C 4000 psi 60 °C 6000 psi 22.5 °C 4000 psi 22.5 °C 50 100 150 TIME (hr) 200 250 300 . . . .. . . . ........... ------ : Fig. 6. Deformation of Models at Different Temperatures, and Stresses, , . .1. : . " , . . . . R L . T lu . . . .. . ORNL-DWG 65-10988 WITH FRICTION REDUCER AT FLOOR AND CEILING VERTICAL SHORTENING OF PILLARS (in.) WITHOUT FRICTION REDUCER WITH FRICTION REDUCER AT CENTER OF PILLAR 0 20 40 60 80 100 120 140 TIME (hr) Deformation of Model Rock Salt Pillars at 4000 psi. Kubota has shown that bedded salt from most deposits will shatter at temper- 'atures between 250° and 380°c, due to the presence of trapped moisture.. A few samples of Kansas salt which were irradjated to exposure doses of 5 x 100 R, or greater, fractured between 260° and 320°c. It is thus concluded that radiation will have little effect on fracturing temperatures. A comparison of thermal conductivities at room temperature of t samples, one control and one irradiated to an exposure dose of 5 x 108 R, showed less than 10% decrease in conductivity due to the radiation. Calculations have been made of the doses resulting from storing future waste Solids in the floor. Integrated salt doses as high as received at distances of more than I ft from the waste containers. Thus it is anticipated that radiation will have little or no affect on the structura), sta-i bility of the mine rooms. There is also a possibility that many of the changes in physical properties which take place upon irradiation at arabient temperatures may be partially an- nealed at the temperatures involved in a waste disposal white crystalline sodium chloride (reagent grade) changes color with increased doses of radiation. At 106 rads it is light tan; it progressively deepens to brown and finally to a dark blue-black_at about 5 : appears to start coloring at around 105 rads.) When salt is irradiated at ele- vated temperatures, the resultant color depends on which process is faster - the accumulation or the annealing of the color centers. At predominates and the salt becomes colored; but at 250°C, the annealing process predominates and the salt remains white. 6 4. RADIATION DOSES AND TEMPERATURE RISES The layout of the experimental area in the demonstration showing the loca- tions of the arrays is shown in Fig. 8. . mical and phosphate glass rod dosimeters have been developed ior measuring the radiation dose received by the salt around the array holes. :: These dosimeters are capable of operating at temperatures up to 250°C with good reliability and accuracy. Measured doses are in reasonable agreement with doses calculated assuming that salt is equivalent to low density concrete as a gamma. absorber. By the end of December 1965 (46 days after start-up of Project Salt Vault, the salt at the surface of the holes had accumulated a dose of about 5 x 107 rads, while doses of greater than 107 rads extended to a depth of about 4 in. To date, there are no apparent significant differences in salt flow be-... · havior between the radioactive and control arrays. was decided that fixed power inputs to each of the array holes would be main. : tained throughout the test. Consequently, the power input to the electrical array is being maintained at 1500 watts/can (10.5 kw for the array), and the radioactive array input power is controlled so that the fission product power plus electrical power totals 1500 watts/can. Fission product power per can at 1 the time of start-up was about 520 watts. It is necessary to add supplementary energy because the fission product power decays rapidly due to the rapid decrease : of activity of the fuei assemblies, because only a small portion of the floor of the room contains heat sources, and because it is desired to reach a peak tem- perature of about 200°C in the limited time involved in the demonstration. :: 11 ... . . . . . . . . . ." · ORNL-DWG 63-774 AR2 , ,- . 4 D 4 2 . OC - .. 10.11 . . 0 0 1 . . 2 . . FEET . - - 50 - 100 150 . .. 21 ! ! - 7 100U 1 - 1 - 1 . 11 11 0 30+1 ELECTRICAL ARRAY 1 IT : 250 30 ft 30ft 09 OS O D 0 +30 . 2004 40ft 60 ft OU . AA 1 . . FEET 118-A-S-4 * HEATED APILLAR O t . 10 109-A-N-21 & MAINE 00 RADIOACTIVE ARRAY II -111-A-U-3 105-A-U-3 20 ft U11 DI . . 115 SOL SPECIALLY MINED AREA 14-ft ABOVE EXISTING FLOOR LEVEL WASTE CHARGING SHAFT FROM SURFACE . O 21 30 ft 11 O OS I : DOO RAMP UP V 1 11 0 D t . C 1 EXISTING MINE WORKINGS SHOWN APPROXIMATE 02 4 FLOOR RADIOACTIVE ARRAY- ::: 2 DU 11 Fig. 8. Layout of Experimental Area. DO .. 1 . 4 . Co- . . y : ! . At the end of December the fuel element temperatures ranged from about 2009 to 250°C, well below the melting point of the aluminum cladding. At that time the salt temperature at the wall of the peripheral holes was about 120°c, and at the wall of the center hole it was about 135°C. In the vertical-center planes of the arrays (about 9 ft below floor level) essentially the entire area within the circle formed by the peripheral heaters (that is, a circle of about 10-ft , diameter) was at or above 100°C. Figure 9 shows the temperature rises as func- tions of time for points in the salt 1 1/2 ft from the centers of two of them array holes (the center hole and one peripheral hole). It may be seen that bot! electrical and radioactive array temperatures are reasonably close to those mated theoretically. It may also be noticed that the temperature in the radio- active array is rising slightly faster than that in the elentrical. This was found to be due to a few per cent drop in input power to the electrical array which has now been corrected. -- Vi.. . 5. SALT MOVEMENT AND STRESS TRANSFER ..;;. 1". . ...n Vertical thermal expansion of the floor in the center of the arrays had reached nearly an inch by the end of December 1965. The floor uplift (measured in feet) at the center, and 10 ft from the center, is shown as a function of time for each array in Fig. 10b. It may be observed that the rate of rise in and near. the array is slowing down. However, at points 25 ft or more distant from the center of the array, the expansion rate does not yet show signs of diminishing ... . ... In Fig. 10a is shown the uplift profiles for room 1 (radioactive array) along the north-south and east-west axes of the room. It is apparent that ther- terial in the floor extends to 40 or 50 ft from the center of the array. The north-south uplift profile shows the restraining effect pro- duced by the presence of the adjacent pillars. Figure 1lb shows the transverse expansion (in inches) taking place in the pillar which borders the radioactive array room on the north side. The curve labeled 109-A-N-2 is a measure of the amount of horizontal expansion taking place between the surface of the pillar (about halfway up from the floor on the south, face of the pillar) and a point about 10 ft in toward the center of the pillar: The curve labeled 118-A-S-l is the corresponding measurement made from the north face of the pillar. (Locations of the gages are shown in Fig. 8.) The po turned on in the array at 806 "standard days," and almost immediately the expan- sion rate on the south side of the pillar increased from about 0.15 in./yr to about 0.45 in./yr, a threefold increase. This marked increase in expansion rate. halfway up the pillar was apparently due in large part to a transfer of thermal stress from the center of the floor, since the increased rate was detected before any appreciable temperature rise had taken place beneath the pillar. The fact that the expansion rate has not tended to decrease with time (as has the rate of floor uplift) is probably due to the fact that some temperature rise is now tak- ing place beneath the pillar (about 10°C at the south edge of the pillar as of December 30). It will be noticed that the north side of the pillar has also ex- perienced a slight increase in stress (curve 118-A-S-1). It should be noted that a similar behavior has been observed in the wall on the south side of the room, and that the behavior of the electrical array room is also essentially the same. . . . The transfer of stress has not been limited to the adjacent pillars, but has extended up to the ceiling as well. That this should be so is attested by the the pillar model tests described earlier. There, it may be remembered, an axially applied Load resulted in movement of salt from the regions above and below the pillar out into the roof and floor regions of the opening (or room). Figure lla shows the movement of the ceiling relative to a point 6 ft up into the roof. The curve labeled 111-A-U-3 is the gage data for a point 10 ft from the 13 ORNL-DWG 66-617- THEORETICAL , 20°C, SALT THERMAL PROPERTIES A CENTRAL HOLE (NO. 4) RADIOACTIVE ARRAY • CENTRAL HOLE (NO. 4) ELECTRICAL ARRAY A PERIPHERAL HOLE (NO.6) RADIOACTIVE ARRAY O PERIPHERAL HOLE (NO.6) ELECTRICAL ARRAY FUEL ASSEMBLY CANISTER INSTALLED TEMPERATURE RISE (°C) 2 5 10 20 50 100 200 .500 1000 TIME (hr) Temperature Rises at 192 ft from Centers of Array Holes. biddi bilo moderne kan 9. : 2014 ORNL-DWG 66-618 (0) 0.075 - FLOOR UPLIFT (ft) g 0.050 7 0.025 & NORTH-SOUTH PROFILE EAST-WEST PROFILE CENTER OF na EXPERIMENTAL AREAL á SOUTHERNT NORTHERN ENTRY TUNNEL EDGE OF ROOM EDGE OF ROOM GoC or num EDC 15 5 5 15 25 35 45 DISTANCE FROM CENTER OF ARRAY (ft) On 0.100 CENTER OF ARRAY FLOOR UPLIFT (ft) 10 ft FROM CENTER ū 0.025 OROOM 1 (RADIOACTIVE ARRAY) •ROOM 4 (ELECTRICAL ARRAY) 0 10 20 30 40 50 60 70 TIME (DAYS) Jigare lola) Floor Uplift Profiles in Room 1 as of December 30, 1965. digue 10(6) Floor Uplift in Array Rooms. ORNL-DWG 66-619 0.15 la) 0.10 105-A-U-3 0.05 LO 149-A-U-3 _0.-oto-0-0-00 STARTUP INCHES Ooo 0.15 (6) 109-A-N-2 0.10 _ -_- 60091 0.05 120-0°448-A-S-1 0_0_0_ab-00-0 STARTUP Oo80 900 650 700 750 800 850 STANDARD DAYS II (0) Movement of Ceiling Relative to Six Feet Up. 11 (6) Transverse Expansion in Pillar 1-2. shine Digme . .. . . . . . . center of the radioactive array (10 ft toward the mouth of the room, Fig. 8). The other curve (105-A-U-3) is for a point about 45 ft from the center of the arrny, in the middle of the access corridor. It will be noticed that almost innediately after start-up (806 "standard days") there was about a fivefold iricrease in the rate of movement near the array (Increase in rate of sag from about 0.12 to 0.55 in. yr), and a substantial increase in the sag rate out in the entry corridor (from izbout 0.16 to 0.41 in./yr). The data for the electri- cul array room show similar movements taking place. Most of this movement ap- pears to be separation of the 2-ft-thick salt layer in the roof from the thin": shale layer immediately above it. Movement rates would be expected to be lower if the salt layer were thick enough to resist buckling. 6. CONCLUSIONS The pillar-model tests have proven useful for understanding the way in which salt movements take place, and it is reasonable to expect that the pre- dictions based on the elevated temperature models will be valid also. No measurable effects of radiation ori the flow of salt were expected, and none have been observed during the first 50 days of operation of Project Salt Vault. Thermal expansion of the floor, and increased transverse expansion rates in the pillars adjacent to the array rooms, have bee The acceleration of movement in the ceiling has exceeded that which was anticipated due to thermal effects in the floor; however, the magnitude of the movement is such that no trouble is expected from this source during the oper- ation of the demonstration. Even in an actual operation, movements of this magnitude (or even greater) should not cause trouble during the time when a ing filled with waste, and, after the room is filled, it is anticipated that it will be backfilled with crushed salt. The increased roof: movement does indicate, however, that an extra margin of stability must be allowed in the initial design of the disposal facility. 1 BIBLIOGRAPHY 2 2 1. S. Serata, "Theory and Model of Underground Opening and Support System," in Proceedings of the Sixth Symposium on Rock Mechanics, Rolla, Missouri, pp. 260-292, University of Missouri at Rolla, 1964. 2. . L. Obert, "Deformation Behavior of Model Pillars Made from Salt, Trona, and Potash Ore," in Proceedings of the Sixth Symposium on Rock Mechanics, Rolla, Missouri, October 1964, pp. 539-560, University of Missouri at Rolla, 1964. 3. T. F. Lomenick and R. L. Bradshaw, "Accelerated Deformation of Rock Salt at Elevated Temperature," Nature 207(4993), pp. 158-159 (July 10, 1965). 4. R. L. Bradshaw, W. J. Boegly, Jr., and F. M. Empson, "Correlation of Con- vergence Measurements in Salt Mines with Laboratory Creep-Test Data," in Proceedings of the Sixth Symposium on Rock Mechanics, Rolla, Missouri, October 1964, pp. 501-514, University of Missouri at Rolla, 1964. 5. B. D. Gunter and F. L. Parker, The Physical Properties of Rock Salt as In- fluenced by Gamma Rays, ORNL-3027 (March 1961). 6. R. L. Bradshaw, F. M. Empson, W. J. Boegly, Jr., H. Kubota, F. L. Parker, and E. G. Struxness, "Properties of Salt Important in Radioactive Waste Disposal," in Proceedings of the International Conference on Saline Deposits, Houston, Texas, November 12-17, 1962, to be published by the Geological Society of America in 1966. 18 . . . . List of Figures : 3 Example of Floor Heave in Older Area of Lyons Mine Example of Ceiling Slab Which has Sagged and Fallen Salt Pillar-Model Detail Fig. 1. Fig. 2. Fig. 3. Fig. 4. Fig. 5. Fig. 6. Pillar-Model Before Testing Section Through Pillar-Model After 60% Deformation Deformation of Models at Different Temperatures and Stresses Deformation of Model Rock Salt Pillars at 4000 psi Fig. 7. Fig. 8. Fig. 9. Layout of Experimental Area Temperature Rises at 1 1/2 Feet from Centers of Array Holes Fig. 10a. Floor Uplift Profiles in Room 1 as of December 30, 1965 Fig. 10b. Floor Uplift in Array Rooms Fig. lla. Movement of Ceiling Relative to Six Feet Up Fig. llb. Transverse Expansion in Pillar 1 and 2 . . . - - -- - - KVETIH - RJ il in END . LU . . 12 - Ko i- : ICA :. . DATE FILMED 4/ 21 / 66 - - C - saung. -. .' LY A VYYSIS