Page images


A major difference between small- and large-scale plant designs is in the process building and its shielded cells. The small plant has a single shielded cell, preferably below grade level, as shown in Fig. 4 (2). But the large plant is in a large building with multiple cells in canyons with one or two processing lines 200-400 ft. long; it also requires large and expensive additions to the cells (such as ventilation tunnels, pipe galleries, access corridors, air-cleaning equipment and remote-disconnect facilities), many of which are either absent or are only minor adjuncts in the small plant. The compactness of the small plant is the principal basis for its strikingly low investment cost compared to the large plant. The cost of either plant type increases as the 0.15– 0.2 power of its capacity; this low scale-up factor has been characteristic of radiochemical operations, but the small-plant capital costs are not on the same curve as the large plants, mainly because of these differences in building design.

EQUIPMENT Although the large plant uses large equipment segregated in separate cells for ready repair by remote techniques without need for emptying and cleaning the entire plant, the unitized aspect of the small plant allows one third to one fourth of the equipment to be removed at a time for external maintenance after certain lines are decontaminated and disconnected. The cell cover is divided into three or four separate sections, and to each are fastened separate units of process equipment,



Generally maintenance problems in radioactive plants are being solved by improved equipment designs, better welding and improved operating techniques. Mechanical devices with moving parts are being replaced by new designsradioactive streams are pumped and metered by air lifts, extraction columns are pulsed by air columns and a minimum number of valves are used. For a small plant we assume 1-gr. life for mechanical equipment and 5-yr. for process equipment.


FIGURE 4.-Small reprocessing plant for power-reactor fuels would have single

shielded cell, preferably below grade level; cell would be -1% X 15 X 30 ft. deep.

Main-process-line equipment is divided into functional groups and installed as unit assemblies. These are further divided into subassemblies of more frequent removal for maintenance is expected. Subassembly status should be given to the solution-metering probes, pulsers, pump and mixing probes, fuel-pin chopper and various in-line instruments (including low-level-radiation monitors, thermalfiow transducers, thermal-conductivity meters, column-feed monitors, solventstream monitors and raffinate monitors).

Maintenance experience on the ICPP Ba140 cell is of interest. After the usual startup problems the process has required only ~1 man-year of mechanical maintenance annually, which is much less than expected by the operating staff. Even so, this cell does not have many improvements we cite in our present smallplant concept.


The large plant is usually big enough to require its own steam plant, exhaust stack, maintenance shop, decontamination facility and "cold" analytical laboratory. The small plant should be able to obtain these facilities jointly with other related industrial operations, thus reducing need for capital outlay but receiving an allocated charge for the service. It is unlikely that a small plant would be located in an isolated area because such an operation probably would be restricted to a “nuclear reservation" where other related processes or plants are installed. Even if these auxiliary facilities must be added to the small-plant capital cost, the increment would be only ~$500,000, or 10% of the capital cost. The steam plant should not be very large for a 65 X 65-ft. building; the exhaust stack would not handle radioactive rare gases because they would be stored.



Most large-plant processes would chemically declad the fuel before recovering uranium or plutonium by solvent extraction in a Purex-like process. However, chop-leach treatment of fuel rods assembled in bundles is being considered for large plants.

In the small plant single fuel pins would be cut into 342-in. lengths and the UO2 leached with nitric acid in a batch-type dissolution. The latest design of Dresden fuel assemblies allows for ready disassembly into individual pins. Other fuels of the long-rod configuration may require special mechanical disassembly before being fed into the pin chopper; chopping rate in the single-pin unit would be adequate for 0.3-tonne-U/day needs.

The choice of a chop-leach and Purex flowsheet combination for Zircaloy or stainless-clad pins of UO, is based on a number of advantages. A minimum number of waste streams and minimum volume of waste is generated by the storage of cladding as metal and the processing of uranium-plutonium by Purex. Because fluorides are not used, voluminous complexing agents do not need to be added to reduce corrosion and the problem of converting waste to solids is simplified. Without chemical decladding off-gas volumes can be kept small. Mechanical removal of cladding eliminates precipitates and

the same process can be used for both stainless-steel- and Zircaloy-clad fuel. The Purex extraction system has proven itself in years of satisfactory operation.


Small-scale plant

Large-scale plant


Combined low and high enrichments, Multipurpose in most cases. Multiple depending on capacity; ss or Zr-2 clad headends for different fuels UO2

[blocks in formation]


Compact, unitized construction

Large for uranium cycles


Direct, with key equipment separately Remote removable


[Steam plant, exhaust stack, maintenance shops, decontamination facilities, cold analytical

labs] Provided at the site, but costs allocated Included in design to the processing plant


Chop-leach followed by 3 cycles of Purex Chemical decladding and dissolution for clad 102; other processes applicable and 3 cycles of Purex; multiple headend

processes ; probably a chop-leach process in the future


Store rare gases; tank storage of neu- Rare gas to stack; tank storage of neutralized liquid waste; jackets stored in tralized liquid waste the liquid waste

We wish to express our appreciation to R. D. Fletcher for his suggestions and discussions on small-scale plant design. This study was conducted under contract AT (10-1)-205 with the Idaho Operations Office of the U.S. AEC.


1. C. H. Stockman. The base load required to attract private industry to the

reactor fuel processing business, Paper presented at the Symposium on Nuclear Fuel Reprocessing, 51st Annual Meeting of the American Institute

of Chemical Engineers (Cincinnati, Ohio, Dec. 7–10, 1958) 2. H. Schneider et al. A study of the feasibility of a small-scale reprocessing

plant for the Dresden nuclear power station, IDO-14521 (1961) 3. T. C. Runion. Private reprocessing of irradiated reactor fuel, in "Proceed

ings of the 1960 Atomic Industrial Forum" 4. 0. Jenne et al. Chapter 600 of Eurochemic Technical Report No. 47 (1959) 5. W. H. Farrow, Jr. Radiochemical separations plant study, Part II design

and cost estimates, DP-566 (March 1961) 6. A. Forcello, J. K. Davidson. Nuclear fuel cycles and the PCUT program, Paper presented at the Congresso Nucleare

Italy, June 13, 1961) 7. AEC reference fuel-processing plant, WASH-743 (Oct., 1957) 8. Costs of nuclear power, TID-8531 (rev.) (Jan., 1961) 9. B. Manowitz. Fuel reprocessing costs, nucleonics 20, No. 2, 60 (1962) 10. D. W. Kuhn, R. D. Walton, Jr. Fuel cycle costs for specific power reactors,

TID-13293 (July, 1961) 11. Chem. Eng. News 34, 5657 (1956) 12. C. G. Manly. Packaged reprocessing plants, Paper presented at the Atomic

Industrial Forum Annual Meeting (Oct. 28, 1957) 13. J. L. Schwennesen, Paper presented at the Symposium on the Reprocessing

of Irradiated Fuels (Brussels, Belgium, May 20–25, 1957) 14. Nucleonics 20, No. 1, 25 (1962) 15. A. L. Ayers, B. M. Legler. Batch processing for kilocurie production of

barium-140, Chem. Eng. Prog. 54, 83 (1958) 16. B. M. Legler et al. Startup operation of a production facility for separating

barium-140 from MTR fuel, IDO-14414 (1957) 17. B. Manowitz, nucleonics 20, No. 2, 60 (1962) 18. R. C. Reid, D. Duffey, J. E. Vivian, nucleonics 14, No. 2, 22 (1956)

[From Reactor Fuel Processing, January 1963)





The question of whether spent nuclear fuels should be processed in large central plants or in small plants associated with a small group of reactors has been a controversial subject for several years. A characteristic of radiochemical operations is that the cost is said to increase approximately as the 0.15 power of the plant capacity. This low scaleup factor is a strong argument in favor of large plants. However, the high cost of transporting fuel to a processing site favors the concept of small, strategically located plants.

The case for small processing plants has been discussed by Slansky and McBride of the Phillips Petroleum Co. in a paper at the 1962 Nuclear Congress and in a recent journal publication. These authors believe that several small plants should be built rather than a single large one. (A license application to build a large plant in western New York State has been filed with the Atomic Energy Commission (AEC) by Nuclear Fuel Services, Inc. (NFS).] * Slansky and McBride indicate that two small plants built by private industry for $5 million each (as compared with $22 to $25 million estimated by NFS for a large plant) could recover all the low-enrichment fuels in the United States for the next 5 to 10 years.

The design features of the small processing plant proposed by a study group of the Phillips Petroleum Co. were discussed in a previous issue of Reactor Fuel Processing. Recent studies 8 indicate that the small plant has an inherent capacity for low-enrichment power fuels up to 0.3 metric ton of uranium per day. This would be the amount of spent fuel from nuclear powerplants of about 750 Mw(e) total capacity.

The compactness of the small plant is the principal basis for its low estimated investment cost compared to the large plant. In the large plant, large equipment is segregated in separate cells for ready repair by remote techniques without the need for emptying and cleaning the entire plant. In contrast the unitized aspect of the small plant is intended to allow one-third to one fourth of the equipment to be removed at one time for external maintenance after certain lines are decontaminated and disconnected. Process equipment is designed to be concentrated in a single shielded cell; this concept significantly reduces expenditures for such things as shielding, ventilation, pipe galleries, corridors, and remote disconnects.

Slansky and McBride point out that, when small batches of fuel are put through a large plant on an independent job basis, the plant holdup and turn-around time raise the unit cost of processing by a large factor. This problem of handling different fuels from many different customers has confronted the Eurochemic processing facility and has reduced its planned operating time to about 150 days per year. Obviously there would be no need to make a small plant fully universal; however, by using the chop-leach head-end method, the plant could process fuel rods clad in either zircaloy or stainless steel.

The estimated processing costs for several processing-plant proposals are compared in table I-1. To put all the designs on a common cost basis, the authors chose a fixed annual charge of 16.9 percent of the capital cost as characteristic of the electricity generating industry. The unit processing cost, as estimated by Slansky and McBride, is remarkably low for the small plant when operating at full capacity ($22 per kilogram of uranium), being only slightly higher than the estimated unit cost for the AEC conceptual plant and the Davison (NFS) plant. If several small-scale processing plants were strategically located with respect to the power-generating industry, it is said that a saving in transportation cost of 0.05 to 0.15 mill per kilowatt-hour might be credited to the processing cost. This saving would correspond to $3 to $10 per kilogram of uranium for Dresden-reactor fuel.

1 Reactor Fuel Processing, 5(4):2 (October 1962).

* C. M. Slansky and J. A. McBride, “Design Philosophy for Small-Size Radiochemical Processing Plants," Preprint Paper No. 81, 1962 Nuclear Congress, New York, June 4-7, 1962.

C. M. Slansky and J. A. McBride, “The Case for Small Reprocessing Plants," Nucleonics, 20(9): 43–47 (September 1962).

Nucleonics, 2019): 24 (September 1962). 5 Reactor Fuel Processing, 5(1): 18–22 (January 1962).


TABLE 1-1.--Summary of cost estimates of processing power-reactor fuels by various plant designs 3


Instantaneous Operating rate

kg of U

per year
per day

Phillips small plant.

Davison Chemical Div.



Du Pont.


AEC Conceptual Plant (USAEC Report WASH-743).

1 Millions of dollars.
* Including fuel fabrication.

3 Actual charge from USAEC Report WASH-743 will average $20 per kilogram of Uranium for a large reactor.

« PreviousContinue »