L-99-003, Provides Response to NRC RAI Re Scale 4.3 Code Validation Rept.Attachment to Ltr Provides Written Response to Latest NRC Rai.Info in Attachment Demonstrates Actions Listed. Correspondence Contains No New Commitments
| ML20205R402 | |
| Person / Time | |
|---|---|
| Site: | 07003089 |
| Issue date: | 04/12/1999 |
| From: | Woolley R UNITED STATES ENRICHMENT CORP. (USEC) |
| To: | Pierson R NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| References | |
| A-058168-01, A-58168-1, L-99-003, L-99-3, TAC-L32076, NUDOCS 9904220330 | |
| Download: ML20205R402 (70) | |
Text
LSEC A cioben.rsy company I
Docket No. 070-03089 April 12,1999 L-99-003 A-058168-01 Mr. Robert C. Pierson, Chief Special Projects Branch Division of Fuel Cycle Safety and Safeguards Office of Nuclear Materials Safety and Safeguards U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555
Subject:
Response to NRC Request for Additional Information regarding SCALE 1.3 Code Validation Report (NRC TAC No. L32076)
The purpose of this letter is to pmvide the United States Enrichment Corporation (USEC) response to an NRC Request for Additional Information (RAI) regarding the SCALE 4.3 code validation report.
In Refere,s;. (1), USEC submitted a nuclear criticality safety report documenting the validation of the SCALE 4.3 computer code system for Atomic Vapor Laser Isotope Separation (AVLIS) facility applications. In Reference (2), the NRC issued an RAI regarding the validation of the SCALE 4.3 code to "detennine the acceptability of the report for the proposed area of applicability." It should be noted that this is the second NRC RAI that USEC has received on this subject. Reference (3) was the first RAI issued by the NRC. On July 21,1998, USEC met M) with the NRC staffin NRC Headquarters and orally presented our responses to the first RAI.
The attachment to this letter provides our written response to the latest NRC RAl. The infonnation in the attachment demonstrates the following:
4 5 the adequacy of SCALE 4.3 to address conditions in the intermediate neutron energy range, including supporting experimental evidence B the adequacy of the biases and margins of safety applied in SCALE 4.3 to extend the range of applicability of the code above 5 weight percent uranium enrichment; up to 10 weight percent E the basis fer a determination that the biases and uncertainties used in the SCALE 4.3 calculation of kerr are independent of multiple parameters and perturbations of those parameters jg g'g 6903 Rockledge Drive. Bethesda, MD 20817-1818 Telephone 301-564-3200 Fax 301-564 3201 http://www.usec.com 9904220330 990412 -
ih. KY Ponsmouth, OH Washington, DC PDR ADOCK 07003089 C
PDR __
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 2 m the basis for a determination that neutron cross-sections for high temperature metals and systems do not need to be considered for the AVLIS application USEC believes that the additional information provided in the July meeting with the NRC and provided in this letter address the all of the technical issues identified by the NRC. As the use of SCALE 4.3 code is important to the completion of the AVLIS facility design and the preparation of the associated license application, we respectfully request that the NRC complete their review and approve of this matter. To facilitate the NRC's review and approval, USEC would be amenable to support another meeting with the NRC to discuss the information contained in this letter. USEC appreciates the staff's effort in this matter.
This correspondence contains no new commitments. Please contact Mr. Jim Slider (703) 620-0769 if you have any questions concerning this matter or desire to have another meeting.
Sincerely,
/
Robert L. Woolley, Manager AVI,lS Nuclear Regulatory Policy and Licensing cc: Mr. Jack R. Davis, Nuclear Regulatory Commission
References:
- 1. Letter from Mr. Robert L. Woolley, USEC, to Mr. Robert C. Pierson, US NRC,"AVLIS Criticality Code Validation Report", dated April 22,1998.
- 2. Letter from Mr. Robert C. Pierson, US NRC, to Mr. Robert Woolley, USEC,"AVLIS Criticality Code Validation Report (TAC No. L32076)", dated December 4,1998.
- 3. Letter from Mr. Robert C. Pierson, US NRC, to Mr. Robert Woolley, USEC,"AVLIS Criticality Code Validation Report (TAC No. L32076)", dated June 10,1998.
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 3 Attachment Response to NRC Request for AdditionalInformation Dated December 4,1998 AVLIS Criticality Code Validation Report (NRC TAC No. L32076)
St.ucture of Response The NRC request for infonnation focuses on questions the staff has about the method USEC used to validate the computer code SCALE 4.3. In particular, the staffinquired about the relationship between data from benchmark experiments and the range of conditions for which USEC seeks approval to use SCALE 4.3.
To put the detailed technical answers into perspective, we have provided an overview of the approach USEC is taking to nuclear criticality safety (NCS) analyses. Detailed responses to the four areas in which the staff seeks additional information are presented in the form of questions and answers, following the overview below.
Overview of AVLIS NCS Approach to Controlled Parameters The NCS analyses of AVLIS Enrichment Plant systems are currently being performed according to general rules for AVLIS criticality safety analyses to be included in the Nuclear Criticality Safety Chapter of the AVLIS license application. These are included in the next two paragraphs.
AVLIS requires that process designs adhere to the double contingency principle. The NCS analyses identify and determine the NCS limits. When determining limits from calculations the credible worst case (or most reactive) combination of special nuclear material density, H/X ratio, solutions concentration, reflection, interaction, interspersed moderation, and measurement uncertainty is considered before nuclear criticality safety limits are established. These conservative bounding assumptions are made at each step in the analysis process. The cumulative effect is to provide a large margin of safety.
In the NCS analysis, a determination is made of the relationship of ken and variations in the controlled parameter. This relationship, along with an assessment of uncertainty for the controlled parameter, and the ability to detect and control process variations that alTect the controlled parameter, is used to establish adequate NCS limits. This gives the analyst an understanding of the sensitivity of kar to changes in the controlled parameters. For each controlled parameter, the values of the parameter that correspond to the Failure and Safety Limit are determined. The Failure Limit is defined as the lowest point at which the system may be critical. The calculation of the Failure Limit includes the uncertainties in the benchmark experiments, the cross section sets, and the computer code system. The Safety Limit is set below the Failure Limit as an added margin of safety (Failure Limit - 0.02). Both the Safety Limit and Failure Limit include appropriate allowances for any bias and uncertainty in data and calculation j
methods used as demonstrated in the validation, and are based on conservative modeling assumptions, as
1 Mr. Robert C. Pierson -
- April 12,1999 -
-L-99-003 Page 4 mentioned earlier. The analyst is responsible for ensuring that the SCALE 4.3 code is validated for the forms, moderation, materials, enrichment, and energy spectra in the bounding safety limit calculations.
Responses to the four specific areas identified in the NRC's December 4,1998 letter follow.
Question (1):
Analysis and review ofprocess conditions occurring in the intermediate energy range and the adequacy ofcurrent experimental evidence to support these cases i
Response
)
To address this question, it is first necessary to define the energy ranges that best describe the AVLIS process:
i Thermal:
E < ~0.3 eV j
Slow Intermediate:
0.3 s E < 10 eV Fast Intermediate:
E 210 eV Since AVLIS is a Low Enriched Uranium (LEU) facility, the neutron slowing-down process naturally populates these basic energy ranges. Most AVLIS systems (in their most reactive credible conditions) involve moderation of neutrons by either water or graphite. In the presence of such moderating materials the neutron energy spectrum has a large thermal energy contingent. Neutrons born from spontaneous fission or external sources have initial energies on the order of at least several MeV. As those neutrons interact with AVLIS materials they either slow down to lower energies, get absorbed, or cause fission. As neutrons slow down towards thermal energies, absorption is greatly increased arotmd 238 the U resonance absorption cross section peak (~ 6eV). Neutrons found in or below the Slow Intermediate range are typically produced in direct scatter from energies that were well above the 23sU resonance absorption range. Therefore the Slow Intermediate range (as defined above) is expected to have very few neutrons for typical 1.EU systems (either from AVLIS or the critical experiments).
The fission fraction produced in each energy group for the validation critical experiments, and calculations typical of all proposed AVLIS processes, was evaluated to clearly illustrate the effects described above. The fission fraction versus energy group was reviewed for each critical experiment.
The maximum fission fraction, among all critical experiments, for each energy group was determined and plotted versus energy group. All current NCS analyses of AVLIS systems were reviewed, and the fission fraction versus energy group for at least 20 calculations of kar from each of the analyses was xamined (over 1700 calculations total). As with the critical experiment calculations, the maximum e
fission fraction versus each energy group was found for AVLIS process analyses. The results of the maximum fission fraction versus energy group for the critical experiment calculations and the AVLIS system analyses are plotted in Figure 1. Figure 1 illustrates the effect of the 23sU resonance absorption region on LEU systems that are parametrically similar to the AVLIS processes. The Slow Intermediate range is shown to have a very small fission fraction for both the critical experiments and the AVLIS I
1
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 5 235 calculations. It is only in the Thermal range where the U fission cross section dominates that fission fraction is able to recover. Fission fraction values in the Fast Intermediate range are primarily affected 235 238 by variations in the U absorp3'U threshold for fast fission. tion and fission cross sections U fission cross section for energies above the Figure 1 illustrates a high degree of similarity between AVLIS systems and the critical experiments even in the Slow Intermediate range where there is little fission activity, it is further noted that SCALE 4.3 uncertainties in that energy region are ofless concern due to the well known 1/v cross sectional behavior of the non-uranium materials ofinterest in the Thermal and Slow Intermediate energy ranges. This well behaved cross sectional behavior in the Slow Intermediate energy range helps to minimize uncertainties in calculations of ker that may be influenced by the few fissions that do occur in that energy region.
Review of Figure 1 illustrates that the fission spectra of the critical experiment calculations provide excellent coverage of the AVLIS systems that have been analyzed to date. In order to ensure those systems are representative of all proposed AVLIS processes, the calculations were categorized and evaluated against a list of actual AVLIS processes and equipment. The following list describes each system (each with at least 20 separate calculations) that was included in the maximum fission fraction analysis described above:
Systems Included
> Refurbishment vacuum systems and vacuum canisters
> Product withdrawal canisters
> Tails withdrawal canisters
> IPD separator pod analyses
> Plant separator pod analyses
> Tails metal disposal analyses
> Tails oxide storage analyses
> Product down-blend furnaces
> Interstage blending furnaces
> Product tundish and mold (includes ingot analysis)
> Interstage tundish and mold
> Uranium Recovery (UR) oxide packaging
> UR pusher furnace
> UR crusher
> UR rotary furnace
> UR graphite packaging
> UR parts transport container
'> UR oxide packaging (columns and container)
> UR vacuum canister handling
> UR parts handling
c l
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 6
> Feed llandling (Fil) 30-bar feed magazine storage
> FII feed bar magazine loading station
> Fil towline conveyer
> Fil 12-pack Nondestructive Assay (NDA) container i
The AVLIS design to date includes a variety of systems and components that have controls to ensure Nuclear Criticality Safety. Those systems and components have been included in Table 1 below, and j
provide a general listing of the AVLIS systems that require NCS analysis. Many of the systems and components listed in Table 1, have already been analyzed, and were directly covered by the calculations described above. Table I systems and components that have not been explicitly analyzed by NCS were j
reviewed for parametric similarities to systems included above. The determination of parametric similarity was based on a qualitative review of the system designs and evaluation of each NCS parameter affecting reactivity for the unanalyzed systems versus the analyzed systems. Table 1 provides a brief discussion under the " Coverage" column to establish whether each item is explicitly covered by the calculations listed above or ifit is covered by parametric similarities to a group of calculations that have already been performed.
Table 1 Systems and Components with NCS Controls System Component Coverage Separator Pod Separator Pod Covered by cases from plant and IPD pod System analyses.
Separator Vacuum Vacuum Vessel Similar to U nugget analysis in plant and IPD pod Vessel analyses.
Vacuum System Similar to U nugget analysis in the canister handling analysis.
i
~ Separator Feed Primary Feeder Similar to feed bar analysis of Fil magazine, System towline conveyer, loading station and NDA container studies.
Secondary Feeder Similar to feed bar analysis of Fil magazine, towline conveyer, loading station and NDA container studies.
Separator Withdrawal Stack Similar to the U metal nugget analyses in the Withdrawal System product withdrawal canister studies.
Withdrawal Covered in Product withdrawal Canister analyses.
l Canisters Feed IIandling Feed Bar Shipping Similar to analysis of feed bar magazine, towline Container conveyer, NDA container, and loading station.
Also only deals with limited quantities of natural assay bars.
Feed Bar Magazine Covered in cases from feed bar magazine storage and Storage analysis.
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 7 Table 1 Systems and Components with NCS Controls System Component Coverage Diverted Feed Bar Similar to cases from feed bar loading station -
Handling Station analysis.
Feed Bar NDA Stub Covered in cases for 12-pack storage analysis and Shipping Container towline conveyer studies.
Tails liandling Tails Withdrawal Covered in the analysis of the tails withdrawal Canister canisters.
Tails Long Term Covered in the analysis of tails metal disposal.
Storage Container Product llandling Product Withdrawal Covered in cases for the product withdrawal Canister canister.
Product Blending Similar to product canister analyses, UR crusher analyses, and product blending furnace analyses.
Charge Make-up Similar to product canister analyses, UR crusher analyses, and product blending furnace analyses.
Product Blending Covered in product down-blend furnace analyses.
Furnace Product Oxide Covered in UR oxide packaging studies.
llandling Container Product Ingot Similar to product tundish and mold analyses.
Shipping Container Interstage Charge Make-up Similar to product canister analyses, UR crusher Processing analyses, and interstage blending fumace analyses.
Interstage Furnace Covered in interstage blending furnace studies.
System Feed Bar Similar to analysis of feed bar magazine, towline Preparation / Cutting conveyer, NDA container, and loading station.
Enriched Bar Covered in the towline conveyer analysis.
Carrier Analytical Canister Sampling Similar to the product canister analysis.
Laboratory Sample Oxidation Similar to the UR oxide packaging analysis, UR pusher furnace analysis, and UR rotary furnace analysis.
Sample Analysis Bound by the optimized, heterogeneous, analysis of the product canisters.
Storage of Samples Bound by the optimized, heterogeneous, analysis of the product canisters.
i i
l
i Mr. Robert C. Pierson 1
April 12,1999 L-99-003 Page 8 Table 1 Systems and Components with NCS Controls System Component Coverage llot Drain Calculations will likely not be performed.
Controls will be used for ensuring always safe concentrations prior to disposal.
Uranium Recovery Pod Parts llandling Covered by UR parts handling.
Loose Films Covered by UR parts handling and UR parts Handling transport container analyses.
Oxidation Furnace Covered by the pusher and rotary furnace analyses.
Oxide Packaging Covered by the oxide packaging analyses.
LLW Packaging and Covered by the graphite packaging analyses.
NDA Melt Ilandling and Similar to product tundish and mold analyses.
Storage Enriched Feed Bar Similar to the analyses of the product and tails Stub Ilandling and withdrawal canisters.
Storage Natural Feed Bar Similar to the analyses of the tails meta disposal.
Stub Ilandling and Storage Pod Transporter Pod Transporter Similar to the plant and IPD pod analyses.
Pod Lag Storage Pod Lag Storage Similar to the plant and IPD pod analyses.
Area Pod Initial Disassembly Similar to the plant and IPD pod analyses.
Disassembly Enclosure Upper Pod Various Similar to the plant and IPD pod analyses.
Disassembly, Workstations Oxidation and Cleaning Thermal Enclosure Various Similar to the plant and IPD pod analyses.
Maintenance Workstations Specific calculations will likely not be performed due to the low credible mass in the thennal enclosure.
Lower Pod Various Similar to the plant and IPD pod analyses.
Refurbishment.
Workstations Final Pod Assembly Various Calculations will not be used. Controls ensuring Workstations no presence of significant uranium will be the method of control.
All Cleaning Vacuum System and Covered in the refurbishment vacuum systems Enclosures Collection Canister and vacuum canister studies.
Mr. Robert C. Pierson April 12,1999 L-99-003.
Page 9 Table 1 Systems and Components with NCS Controls System Component Coverage OITNormal OffNormal Similar to the plant and IPD pod analyses.
Maintenance Maintenance Enclosure Central Waste TBD Similar or bounded by the various metal and Management oxide cases referenced above.
Table 1 establishes that all proposed AVLIS systems and components that require NCS analyses either have been explicitly covered by the calculations listed previously, or are parametrically similar to the systems previously analyzed. Therefore, the approximately 1700 AVLIS system calculations that were included in the determination of maximum fission fraction versus energy group adequately represent AVLIS processes.
Previous Analyses of the Intermediate Range Previous information provided to the NRC regarding the " intermediate energy range" was provided as a discussion paper presented by USEC during a July 21,1998 meeting with the staff. This paper was j
documented in the NRC memorandum "Afecting Summaryfor Afecting with U.S. Enrichment Corporation to Discuss Criticality Safety Validation Report, " datedJuly 23,1998,from Drew Persinko to Robert Pierson, which summarized the meeting. That analysis evaluated the energy spectra using the value of Energy Corresponding to the Average Lethargy Causing Fission (ECALCF) produced in the SCALE 4.3 output. ECALCF provides an estimate of the average lethargy causing fission. As Figure 1 demonstrates, some AVLIS processes include a double peak in fission fraction with significant fission activity in both the Thermal and the Fast Intermediate ranges. ECALCF values act to average that behavior. Since the previous definition ofIntermediate energy included energies from about 0.625 eV to about 100 kev, it was reported that both AVLIS systems and some of the critical experiments were of intermediate energy.
At the time of the July 21,1998 meeting with NRC, there was still an open question about whether AVLIS had any intermediate energy systems. Since that time, considerable analysis work has been done on AVLIS systems. This response acts to clarify the previous response by demonstrating that both AVLIS systems and the critical experiments include significant Thermal and Fast Intermediate behavior, but neither AVLIS nor the critical experiments include significant Slow Intermediate fission spectra.
- Conclusions The analysis shown in Figure 1 shows a high degree of similarity between AVLIS systems and the critical experiments chosen for validation. Further, the analyzed cases adequately represent AVLIS systems since all systems with fissile material were included directly in this study or indirectly by i
parametric similarities with other systems. The excellent agreement in maximum fission fraction per energy group between the critical experiments and the AVLIS analyses shows that AVLIS processes are well covered by the critical experiments.
a
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 10
. Question (2):
Application ofappropriate bias and margins ofsafetyfor the extension ofenrichment values in the range of5 - 10 weightpercent
Response
As previously provided in, " Discussion Material on the NRC 's Requestfor AdditionalInformation Letter, Dated April 22,1998, A VUS Enrichment Plant,for July 21 Meeting with the NRC, the appropriateness of the 0.02 margin of sub-criticality has been established. That information shows the critical experiment data set sufficiently covers the range of applicability for AVLIS criticality calculations, supporting operations and off normal conditions as they are currently defined. That conclusion is further supported in the response to Question (1) and (3).
The application of appropriate bias and margin of safety is addressed in this response by comparison of the statistical methodology used in the AVLIS validation versus the approach suggested in NUREG/CR 6361. This discussion will demonstrate that the AVLIS statistical methods, plus a 0.02 administrative margin, result in a conservative margin of safety compared to the NUREG methodology. The AVLIS NCS program incorporates the use of parameter studies into the development oflimits on controlled parameters (e.g., Reference 5). These parameter studies allow limits to be set such that no single credible process upset will result in exceeding the safety limit established in the AVLIS validation report. Considering the AVLIS validation margin of safety and the sensitivity based approach of the AVLIS NCS program, the application of bias and margin of safety in the AVLIS program is shown to remain conservative and adequate.
Since there is no simple correlation between k,n and variations in physical parameters, the safety of AVLIS operations where reactivity is calculated is based on an understanding of the safety margins provided by controlled parameters. For each controlled parameter, a determination is made using parameter studies to determine the relationship between keg and variations in the controlled parameter.
)
This relationship is used to establish adequate safety margins. This approach shifts the focus from an arbitrary ken alue as an indication of the available safety margin, to an understanding of the sensitivity v
of ken to changes in controlled parameters.
For each controlled parameter, the values of the parameter that correspond to the Failure and Safety Limits are determined. The Failure Limit is defined as the lowest point at which the system may be critical based on the uncertainties in the benchmark experiments, the cross section sets, and the computer code ~ system. Its calculated ken alue plus uncertainties therefore is the Lower Tolerance v
Limit value as seen in the validation report (0.9753 for the 238-group cross section seR The Safety Limit is set below the Failure Limit value as an added margin of safety. The Safety Liuit either does not exceed a ken of LTL-0.02 (0.9553 for the 238-group cross section set) or 85% of the Failure Limit value when measured in terms of a controlled parameter. Either of these limits may be used
Mr. Robert C. Pierson April 12,1999 L-99-003 Page11 independently to define the Safety Limit. AVLIS operations will ensure that no single failure will result in exceeding the Safety Limit value for any given parameter.
A Safety Limit of(LTL-0.02) provides a. sufficient margic of safety because systems with enrichments lower than 10 wt. % U235 are generally less sensitive to changes in reactivity parameters than more enriched systems. As an example, Figure 2 shows the results of ken ersus Fraction of Critical Mass for v
10 wt. % enriched uranium. This system is typical of uranium metal accumulating in the separator or collection canisters and approaches criticality slowly and nonlinearly. The non-linearity of kenwith mass is significant (ken ~ FRAC"). The fraction of critical mass at a 0.02 margin is approximately 84%, and at a 0.05 margin is approximately 65%. Failure to account for non-linearity of the relationshy between ken and fraction of critical mass is one of the inherent shortcomings of any criticality safety l
regulation that prescribes a rigid limiting value of ken. There is little added safety, and there are significant processing and economic drawbacks in going from a margin of 0.02 to 0.05 for these nonlinear systems.
As supporting information for the selection of 0.02 margin of sub-criticality, a statistical analysis comparable to that found in NUREG/CR-6361 was performed. In NUREG/CR-6361, a statistical I
method for establishing an Upper Safety Limit (USL) is described. This USL is analogous to the Recommended Safety Limit described in the AVLIS Validation Report. A statistical analysis code, USLSTATS provided by ORNL, facilitates the analysis described in NUREG/CR-6361. This code determines a linear regression fit between the calculated ken and the parameter ofinterest (e.g., ll/X, fission energy, enrichment, etc.). The code will establish an USL using a user-supplied administrative margin and one using a closed interval approach. The closed interval method establishes the margin of sub-criticality as the minimum difference between the confidence interval and the tolerance limit. This definition is unique to NUREG/CR-6361 and does not appear in any cor.sensus criticality safety standard.
NUREG/CR-6361 presents a methodology to determine the adequacy of a selected arbitrary margin of reactivity. This methodology represents the minimum margin of sub-criticality as the minimum difference between the confidence limit and the tolerance limit. Using this methodology, the USL was i
calculated for each of the three cross section libraries versus enrichment, H/X, and Energy Corresponding to Average Lethargy Causing Fission (ECALCF) using 0.95/0.99 confidence and an administrative margin of 0.02. The calculated minimum margin of sub-criticality for the 238-group cross section library is below 0.014, for the 27-group cross section library is between 0.015 and 0.017, and for the 16 group is between 0.017 and 0.022 (Figure 3). The one set that does not have a calculated minimum margin less than 0.02 is the 16-group cross section library when trended against ECALCF.
The difference is due to the variance in the data for this library.
The Recommended Safety Limit in the AVLIS validation report was developed using a different statistical approach. The calculational reedts were tested to determine if trends existed. Linear least square fits were determined for selected key parameters (II/X, ECALCF, enrichment, etc.). The i
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 12-2 correlation coefficient (R ) was determined for these fits. Ifit was below 0.5, no trend was assumed to exist. For the parameters ofinterest, no significant trends were evident. The Validation Report shows the data to be normally distributed. The Lower Tolerance Limit (LTL) was developed for a 0.95/0.99 percent confidence level. The pooled variance used in establishing the LTL was the sum of the variance -
about the mean, the variance of the calculations and the variance of the experimental uncertainties. The Recommended Safety Limit was then established as the LTL minus the margin of sub-criticality (0.02).
In comparing the USL and the Recommended Safety Limit, the Recommended Safety Limit is lower than the USL in all cases (see Figure 5). The Recommended Safety Limit is the Lower Tolerance Limit (LTL) minus the margin of sub-criticality. 'Since the Recommended Safety Limit is less than the USL, the AVLIS method of validation is more conservative than the method specified in NUREG/CR-6361.
The USL and the Recommended Safety Limit for each of the three parameters (enrichment, li/X, and ECALCF) and the three cross section libraries are shown in Figures 5-13.
Question (3):
Review ofmultiple parameter interdependence andperturbationfor those parameters specific to the A VLISprocess.
Response
To clarify the intent of Question 3, telephone discussions between Dr. Ron Koopman (LLNL) and Mr.
Jack Davis (NRC AVLIS project manager) were held on 1/21/99 and 2/3/99. Those discussions led to i
the identification of the following items to be addressed in this response for Question 3:
a) Additional discussion is needed regarding which reactivityparameters are usedfor A VLIS NCS control or are characteristic ofthe A VLISprocesses.
b) Additionalinformation is needed to demonstrate that the A VLIS critical experiments cover the range
- ofapplicabilityfor each important parameter.
c) Regarding the interdependence ofparameters, the NRC acknowledges that a complete set ofmulti-parameter critical experiments is not available, but expects unequivocal demonstration that our critical experiments cover the range ofapplicability.
This study demonstrates that bias and uncertainties in SCALE 4.3 calculations of critical experiment kerr values are independent of combinations of the NCS control parameters within the AVLIS range of applicability. Bias and uncertainties in calculations of ketr are also shown to be independent of the energy spectra that are produced in the AVLIS range of applicability. Therefore, the statistical methods for calculating the safety and failure limits established in A VLIS Nuclear Criticality Safety Report, Validation ofSCALE 4.3, remain conservative and the AVLIS calculations of safety and failure limits
- remain conservative over the entire range of applicability.
L Mr. Robert C. Pierson April 12,1999 L-99-003 Page 13 Question 3(a) Response:
He AVLIS NCS program provides the nuclear criticality safety basis for operation of the AVLIS processes. The program ensures nuclear criticality safety by implementing the double contingency principle to ensure that parameters affecting reactivity are controlled within established limits or to ensure they have no credible unsafe values. Nuclear criticality safety is maintained by utilizing robust methods of control for the parameters affecting reactivity within limits where changes in the limited parameter are detectable and controllable below LCO values.
The following parameters affect reactivity for AVLIS processes:
. Geometry Spacing Volume Fixed Neutron Absorber Piece Count Mass Moderation Concentration Material Specification Uranium Enrichment Soluble Neutron Absorber Reflector NCS analysis of AVLIS processes defines controls based on these parameters to ensure that credible unsafe values are not reached. On a case by case basis each analysis evaluates a system over the range of applicability, which is established within the values found in the validation critical experiments. No extensions to the range of applicability beyond the limiting values of the critical experiments are made in the analysis of AVLIS processes.
Question 3(b) Response:
The critical experiments used in the validation were chosen because they featured parametric similarities to the process parameters found in AVLIS Enrichment Plant systems. The list of critical experiments used in the validation, and the critical parameters covered by those experiments, is provided in Table 2.
The critical experiments used in the AVLIS validation were chosen because they developed a range of l
applicability for each parameter that enveloped the range of parameter values expected in the AVLIS processes. These experiments establish the full range of each parameter affecting reactivity. The AVLIS area of applicability is established within those limiting values of each parameter bounded by
. the critical experiments. Each analysis must remain within the range of applicability for each parameter to conservatively bound the AVLIS system being analyzed. No extensions to the range of applicability 4
beyond the limiting values of the validation critical experiments are made in the analysis of AVLIS processes.
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 14 Table 2 AVLIS Parameters Covered by Benchmark Critical Experiments Reference Description of Parameter ofInterest Configurations Document Experimental Configuration Modeled Compound Experiments BAW.1484-7 Nine fuel rod clusters simulating Uranium Form (UO2) 34 LWR fuel elements; grouped in a Enrichment (2.46 w/o) 3 x 3 array;& moderated &
Other Materials (borated water, redected by borated water. Fuel B4C pins, stainless steel) rods are aluminum clad 2.46%
Spectrum (thermal) enriched 0.406 inch diameter Moderation (water)
UO. Rod OD is 0.475 inch and lieterogeneity (different spacing 2
active length is 60.4 inches. Rod between groups of fuel rods) pitch within a cluster is 0.644 inches.
IEU-COMP-TitERM-001 U(30)Fc(CF ),, cubes and Enrichment (30 w/o) 3 2
polyethylene cubes stacked with Other Material (polyethylene, a paraffin redector.
paraffin)
Spectrum (thermal)
Moderation (polyethylene)
LEU <OMP-TilERM Ol8 U(7)O rods (0.743 cm OD,69.04 Uranium Form (UO2)
I 2
cm II) clad in stainless steel Enrichment (7 w/a)
(0.7792 cm ID,0.8324 cm OD)in Other Materials (stainless steel) a square pitch (1.32 cm) lattice Spectrum (thermal) reDected and moderated by water.
Moderation (water) lieterogeneity (square pin pitch)
LEU-COMP-TIIERM-019 U(5)O rods (0.436 cm OD,59.66 Uranium Form (UO2) 3 2
cm II) clad in stainless steel Enrichment (5 w.o)
(0.440 cm ID,0.500 cm OD)in a Other Materials (stainless steel) hexagonal pitch lattice reuected Spectrum (thermal) and moderated by water.
Moderation (water)
{
lieterogeneity (hexagonal pin pitch) t EU-COMP-TilERM.020 U(5)O rods ( 0.460 cm OD,59.66 Uranium Form (UO2) 7 2
l cm 11) clad in zirconium alloy Enrichment (5 w/o)
(0.460 cm ID,0.610 cm OD) in a Other Materials (zirconium) hexagonal pitch (1.30 cm) lattice Spectrum (thermal) reDected and moderated by water.
Moderation (water)
Criticality was established by lieterogeneity (hexagonal pin water height.
pitch) l l
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 15 Table 2 AVLIS Parameters Covered by Benchmark Critical Experiments Reference Description of Parameter ofInterest Configurations Document Experimental Configuration Modeled 1.EU-COMP-TilERM 021 U(5)O rods (0.460cm OD,59.66 Uranium Form (UO2) 6 2
cm II) clad in zirconium alloy Enrichment (5 w/o)
(0.460 cm ID,0.610 cm OD)in a Other Materials (zirconium, hexagonal pitch lattice reflected borated water) and moderated by borated water.
Spectrum (thermal)
Criticality was established by Moderation (water) water height.
lieterogeneity (hexagonal pin pitch) 1.EU-COMP TilERM 022 U(10)O rods (0.416 cm OD,85.6 Uranium Form (UO2) 7 2
cm 11) clad in stainless steel alloy Enrichment (10 w/o)
(0.430 cm ID,0.510 cm OD)in a Other Materials (stainless steel) i hexagonal pitch lattice reflected Spectrum (thermal)
{
and moderated by water.
Moderation (water) 11eterogeneity (hexagonal pin pitch)
LEU-COMP-TiiERM-023 U(10)O rods (0.416cm OD,85.6 Uranium Form (UO2)
G 2
cm II) clad in stainless steel alloy Enrichment (5 w/o)
(0.430 cm ID,0.510 cm OD)in a Other Materials (stainless steel) hexagonal pitch (1.40 cm) lattice Spectrum (thermal) reflected and moderated by water.
Moderation (water)
Criticality was established by lieterogeneity (hexagonal pin water height.
pitch) 1.EU-COMP TilERM-024 (t(10)O rods (0.416 cm OD,85.6 Uranium Form (UO2) 2
)
2 cm II) clad in stainless steelalloy Enrichment (5 w/o)
(0.430 cm ID,0.510 cm OD)in a Other Materials (stainless steel) square pitch lattice reflected and Spectrum (thermal) moderated by water.
Moderation (water) lieterogeneity (square pin pitch)
LEU-COMP TilERM-025 U(7.5)O rods (0.416 cm OD, Uranium Form (UO2) 4 2
85.6 cm II) clad in stainless steel Enrichment (7.5 w/o)
(0.430 cm ID,0.510 cm OD)in a Other Materials (stainless steel) hexagonal pitch lattice rellected Spectrum (thermal) and moderated by water.
Moderation (water) lieterogeneity (hexagonal pin pitch)
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 16 Table 2 AVLIS Parameters Covered by Benchmark Critical Experiments Reference Description of Parameter ofInterest Configurations Document Experimental Configuration Modeled LEU-COMP TilERM-026 U(4.92)O rods (0.145 cm ID, Uranium Form (UO2) 4 2
0.753 cm OD,60.4 cm 11) clad in Enrichment (5 w/o) zirconium alloy (0.785 cm ID, Other Materials (zirconium) 0.915 cm OD)in a hexagonal Spectrum (thermal) pitch lattice reflected and Moderation (water) moderated by water.
lieterogeneity (hexagonal pin pitch)
Total Compound Experiments 77 Metal Experiments IEU-MET-FAST-001 Stacked IIEU metal disks (10.50 Uranium Form (metal) 4 in OD,0.800 in thick) and natural Enrichment (36,56 w/o) uranium metal disks (10.50 in Other Material (natural uranium)
OD,0.600 in thick) with average Spectrum (fast) enrichment of between 36 and 56 wt.%.
IEU-MET-FAST 002 Stacked pairs ofIIEU metal disks Uranium Form (metal) 2 (15.0 in OD,0.1195 in. thick) and Enrichment 16.1%
natural uranium metal disks (15.0 Other Material (natural uranium) in. OD,0.589 in. thick). He core Spectrum (fast) is reflected by natural uranium (3.00 in. base,2.96 in. top,2.99 in, sides). Critical number of pairs is 17.57.
IEU-MET-FAST-003 Bare U(36) concentric spherical Uranium Form (metal)
I shells of metal (2.8 cm ID,30.6 Spectrum (fast) em OD).
IEU-MET rAST-004 U(36) concentric spherical shells Uranium Form (metal)
I of metal (4.0 cm ID,28.0 cm OD)
Other Materials (graphite) reflected by a graphite shell(28.0 Spectrum (fast) em ID,34.4 cm OD).
IEU-MET FAST-005 U(36) concentric sphericalshells Uranium Form (metal)
I ofmetal(4.0 cm ID,26.5 cm OD)
Other Materials (steel) reflected by steel shells (26.5 cm Spectrum (fast)
ID,43.0 cm OD).
i Mr. Robert C. Pierson
)
April 12,1999
{
L-99-003 "E"
.rable 2 AVLIS Parameters Covered by Benchmark Critical Experiments Reference Description of Parameter ofInterest Configurations Document Experimental Configuration Modeled Y-DR-81 U(92.3) metal cylinders and Uranium Form (metal) 1I annuli both bare and reflected.
Other Materials (polyethylene, (Tables 2 and 6 are polyethylene graphite) reflected; Table 4 is graphite Spectrum (mixed, fast and reflected.)
thennal peaks)
{
Moderation (polyethylene, graphite)
ZPR6 (U9)
U(9) composed of enriched Uranium Form (metal) 3 uranium (U(20), U(93)) plates Enrichment (9 w/o) along with natural and depleted Other Materials (depleted and plates.
natural uranium)
Spectrum (fast)
Total Metal Experinwnts 23 Solution Experiments LEU-Sol-TilERM.003 U(10)O NO solution in bare Uranium Fonn (uranyl nitrate) 9 2
3 spheres (66 cm,88 cm,120 cm Enrichment (10 w/o)
ID)either panially filled or fully Spectrum (thermal) filled.
Moderation (water) oRNL-2968 U(4.89)O F solution in a 20 inch Enrichment (4.89 w/o) 6 2 2 diameter stainless steel (1/16 Other Materials (stainless steel) inch)cylinderboth bare and Spectrum (thermal) rcDected.
Moderation (water) sTACY U(9.97)O NO solution in a steel Uranium Form (uranyl nitrate) 12 2
3 cylinder (59.5 cm ID) both bare Enrichment (9.97 w/o) and water reDected.
Other Materials (stainless steel)
Spectrum (thermal)
Moderation (water)
Total Solution Esperiments 27
)
4
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 18 Question 3(c) Response:
The basic process of selecting critical experiments that are similar in nature to processes found in AVLIS, leads to a significant degree of correlation of data between parameters. For instance, mass and volume are both NCS control parameters, and will clearly demonstrate significant correlation among similar critical systems. Mass and enrichment, mass and poison concentration, and all other combinations of parameters must be correlated in any set of similar l
critical experiments. This does not affect the integrity of the validation process. It will be shown below that such inter-dependence does not adversely affect the bias and uncertainties in calculations of kenmade by SCALE 4.3, within the AVLIS range of applicability.
Eperuv Spectrum and the NCS Control Parameters:
As described below, the NCS control parameters can be consolidated into as few as 10 separate parameters, each of which covers a wide range of possible values (~ 40 paramete: sub-groups are evaluated below for the AVLIS range of applicability). There are insufficient benchmark data to support evaluating the effect of multi-parameter inter-dependence by evaluating ken ersus all v
combinations of values for all NCS parameters. With hundreds of parameter combinations possible, the benchmark are too sparse to draw statistically significant conclusions. In addition, most of the possible combinations are irrelevant to evaluation of the AVLIS process.
An alternative approach is to evaluate ken versus the energy of neutrons causing fission for each parameter sub-group. Parameters influence ken by affecting the number of neutrons available to cause fission at any given energy. The neutron energy distribution is established as a result of the energy dependent cross sections for the AVLIS materials that are present. Thus, the energy of neutrons causing fission is the fundamental parameter needed for examining the effects of multiple parameters within the AVLIS area of applicability. It will be shown that within the range of applicability for each NCS parameter:
> The critical experiments cover the entire range of possible spectra, from thermal to fast, over the range of applicability for each NCS parameter.
> There are no significant trends in ken ersus energy for any parameter subgroup.
v In other words, SCALE 4.3 calculations of keg maintain their accuracy over the wide variety of neutron energy spectra, which is a direct result of the AVLIS range of applicability. Figures 14-49 present plots of ken versus thermal fission fraction for each of the parameter sets important to the AVLIS area of applicability. For instance, the range of applicability for II/X is covered with critical experiments ranging from themial to fast energy spectra and shows no identifiable trend in ken versus energy. This means that the bias and uncertainty of the SCALE 4.3 calculation does not change because of the spectral changes that are induced by varying H/X.
Thermal Fission Fraction:
In Figures 14-49, the dependence of ken on energy spectra for all critical experiments within a given parameter sub-group, was evaluated by plotting k,n versus the thermal fission fraction.
I
. Mr. Robert C. Pierson April 12,1999 L-99-003 Page 19 The thermal fission fraction value for each critical experiment was calculated by summing the l
fission fraction per energy group (part of the KENO V.a output) for energies below 0.1 eV. This analysis uses 0.1 eV as the cutoff for thermal energies since it is well above the 0.0253 eV,2200 l
m/s, cross r,ection and well below the 0.3 eV cutoff for Slow Intermediate neutron energies (as l
described in the response to Question 1). Unlike ECALCF, which includes results from the l
entire energy range, the thermal fission fraction is independent of double fission fraction peaks in the fast and thermal energy bands. Averaging the spectral behavior of a double peaked system can lead to misleading conclusions about the spectral characteristics of the system. Therefore, the thermal fission fraction is the integrated fission fraction over thermal energy groups and provides excellent insight into "how thermal" the system ofinterest behaved. Since AVLIS systems are i
dominant in the Thermal and/or Fast Intermediate energy ranges, the thermal fission fraction
- plots will show thermal systems to have thermal fission fraction values at or above about 0.5, l
while fast systems will have thermal fission fraction values near zero. Few systems have thermal fission fraction values between 0.1 and 0.5, which is a byproduct of the fission fraction versus energy results discussed in Question 1.
Parameters:
.. In' order to evaluate how each parameter affects the energy spectrum, the range of values for each parameter in the AVLIS critical experiments was subdivided. The parameters listed in the
- Response to Question 3 (a) were consolidated based on the neutronic parameter being controlled:
-Geometry
-Spacing
-Volume -
-Neutron Absorbers
-Mass
-Interaction
-Moderation
-Heterogeneity (includes concentration, and density affects)
-Enrichment
-Reflection The range of parameters covered by the critical experiments was then further divided into l
appropriate sub-groups. The calculated ker for each critical experiment within a parameter sub-group was plotted versus its thermal fission fraction. The resulting plots, one for each parameter sub-group, are shown in Figures 14-49. These plots show how each sub-group affects the energy spectrum and kor. The sub-groups were selected as follows:
NCS Parameter Parameter Sub-groups Geometry Cylindrical, Spherical, Cuboid, Single Unit, Array Spacing Homogeneous, Lattice Cell l
Volume Vol. > lx 10' ce; lx 10' cc < Vol.< 1x 10' cc j
Neutron Absorbers Boron, Oxygen, Stainless Steel, Iron, Aluminum Mass Mass > lx 10" g; lx 10' g < Mass < lx 10 g, 5
Mass < 1x 10 g, Interaction Single Unit, Anay Pitch / Fuel OD between 1 and 2 1.
p Mr. Robert C. Pierson April 12,1999 L-99-003 l-Page 20 NCS Parameter Parameter Sub-groups Pitch / Fuel OD between 2 and 3 l
Pitch / Fuel OD between 4 and 5 l
Moderation II/X < 100; 100 < ll/X <500; il/X >500 graphite, water, polyethylene, paraffin lieterogeneity (includes Compound, Solution, Metal, llomogeneous, Lattice Cell, concentration, and Pitch / Fuel OD between I and 2, density affects)
Pitch / Fuel OD between 2 and 3, Pitch / Fuel OD between 4 and 5 Enrichment Assay < 5%; 5%< Assay <10%; 10%< Assay <60%; Assay >60%
Reflection borated water, water, paraflin, polyethylene, steel, nickel, natural uranium, bare Trends:
Trends in ken ersus thermal fission fraction would indicate a systematic, energy dependent bias v
in the ability of SCALE 4.3, with energy dependent cross sections, to calculate particular systems. Since no trends have been identified, a single conservative bias, such as we have used, is appropriate. A goal of validation is to identify trends (if any exist), or to verify that the calculations of ken show no trends with the important NCS parameters within the range of applicability.
Results The sub-group plots (provided in Figures 14-49) of ken versus thermal fission fraction were reviewed, leading to the following conclusions:
k For each parameter, the sub-group plots provided data, which adequately covered the full range of thermal fission fractions. Metal-only systems with obviously Fast Intermediate spectra, did not include higher values of thermal fission fraction, since those spectra are almost entirely comprised of fast neutrons.
>. No trends were identified in ken versus thermal fission fraction for any of the subgroups.
> Each sub-group yielded kenvalues reasonably distributed about the overall average ken value.
>. Each sub-group yielded ken alues that were well above the AVLIS failure limit.
v I-
==
Conclusions:==
l The energy of neutrons causing fission has been selected as the fundamental parameter for I
examining the effects of other parameters within the AVLIS area of applicability.
The experiments selected cover the full range ofimportant parameters within the AVLIS area of applicability. The effects of parameter interdependence for AVLIS were evaluated by plotting ken versus the thermal fission fraction for all critical experiments within a given parameter sub-l
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 21 group. None of the plots of ken ersus thermal fission fraction, for any of the parameter sub-v groups, showed significant trends. The plots for each parameter sub-group yielded ken alues v
reasonably distributed around the overall average k<n for all critical experiments.
' Bias and uncertainties in the calculations of ken have been shown to be independent of the possible energy spectra that are produced in the AVLIS range of applicability. Therefore, the statistical methods for calculating the safety and failure limits established in A VLISNuclear Criticality Safety Report, Validation ofSCALE 4.3, remain valid and conservative.
Question (4):
Adequacy ofavailable cross-sectionsfor high temperature metals and system that are part ofthe A VLISprocess or supporting rationale that such conditions are not possible.
Response
During normal separator operations, uranium is fed into the separator continuously where it is melted, vaporized and condensed such that it exists mostly as flowing molten metal which is then cast into nugget form and dropped into canisters attached to the separator. These canisters are weighed continuously and removed when full. The amount of uranium allowed to be held-up in the separator is measured and strictly limited to an amount which is highly sub-critical under all normal _ and abnormal conditions which do not involve water. The hold-up limit for the separator is based on a postulated accident involving borated cooling water leaking into the separator and forming an optimized mixture with the enriched uranium, which is then reflected by graphite from the collapse of the pod structure. This highly conservative scenario is used to set the operating limits for the separator.
If a leak were to occur in a hot separator, the borated water would rapidly turn to steam, which would be removed by the vacuum system or go out the pressure relief valves, if the vacuum system shut down. Enrichment would cease and the liquid uranium films would solidify and oxidize. The system would continue to be highly sub-critical until it cooled enough for liquid water to accumulate. Under these conditions the SCALE 4.3 code system will not have to make accurate calculations of high temperature uranium ' systems. Calculations of ken under these conditions would be at temperatures close to the boiling point of water, where cross sections are well known and validation benchmarks have been obtained (e.g. the Russian benchmark experiments include enrichments between 5 and 10% and hot borated water).
I
v( ha O h'8
>1 ~
eS r5b"
$O 1
\\
d\\
2 3
/ V 4
^
5
\\
s
\\
s u
6 l
c i h,
\\
a i
c A
7 ti i l
C c l
r S
)
8 d U V
na TA 9
se 10 s 2 0
a C,
1 1
m i\\
r u m AA p
1 1
//
V o
t 2
i r
f G p 1
1 y a l
~
lf 4
3 p
g g l
1 1
p u
r h e
e x
r n e a
Q 4
jG u
E r 1
p g
f h
i m
u F
c 5
a 1
E t
E rr r f 6
^
o m 1
fn s 7
i o
f 1
m ite m
8 mi 1
F m Ve 9
w n n 3
1 o r 0
i is t
0 s s 2
2 F d 2
i p
t md a
1 n
s 2
mim ids G
i u
c 2
t a
2 l
i ix <
Oc
\\
a n
3 MO
~ "-
2
\\
a 4
2 Y
5 2
\\
y\\
62 7
2 5
o 3
0
g. xoXa n 34e
>5 '
$c E
r6* gw 9
1 Tsa y J
!i j
1iI fi1
!l 8
w 1
1 7
1 6
lI iI:
1 4t 5
1 4
l!
l!
tb 1
~
3 1
ss M
2 a
1 4!
i la s
c yl 4!I 1
s i
1 s
t A
i i
r l
l C
ac f
- l 1
i o
tir n
C 2
o f
9 o i
g t
l!
0 n c
.iF a
o r
B F
c 8
a
,l d r
0 F sv ev 7
- lI i
0 tce f
6 f
i!
0 e
-k 5
0 4
0 3
m I
0 s
a.
i 2
0 a
m 1
a i
0 m
0 2
1 1
9 8
7 6
5 4
3 2
1 0
1 1
0 0
0 0
0 O
0 0
0 m
Ih1 m
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 24
.i b xEd 2 di 8 mao I
EE0'O N
6LO
]
LLO'O 4
9 e
g.j
}
S LO'O 8
th0'0 u)
E 2
tLO h
h h
3 3
N R
3 8
o o
o o
o o
c:
o o
o o
o c.
Ai!!eo!)poqns 6
- o u! Jen wnm!u!w peneinoleo
Mr. Robert C. Pierson April 12,1999 L-99-003 Page 25 u.
O x c<
2EO EEO I
ZI
- I I
CLIYO LO*0 l
D
!!B Z LO'O J
To 200'0 b-it N
o 900'0 t/)
em O
LO O w
n a
w w
w n
c.
r r
e o
o o
o c
q q
q q
o q
q c-l o
o o
o o
o o
e j
11m!14ejes popuemmoces j
snu!w 1sn Jo1 en!eA wnm!u!n 3
)
K :' :c X a O 7nga l
o L
f
>7=
$c eSL f
k US re* gw
?
- ==a y 06 0
5 0
4
+
g+
)
p o
l u
w o
(
r t
G n
5 8
0 e 3
g e
3 m 2
h ru r
c g
o i
r n
f iF E
'e m
h c
i rn E
02 e
+
0 P'
^
1
' ^
- i I ;
9 0
3 2
1 0
9 8
7 6
5 0
0 0
0 9
9 9
9 9
1 1
1 1
0 0
0 0
0 ej
K' so?a O6WC8 e
i 5
>7= P 8*
L r6*o8 n S h e U?
k o
$a t*
5
?_
"t 0
0 6
1 00 4
1 002
+
1 0
0 p
0
+.
u 1
o
+
r 6
G
- +
e 8
r 3
u 2
+
g 0 X i
r 0 /
F o
8 H d
f X
I H
006 3*
0 04
+
00 2
s*
+
o><1 o *llI **3 4
0 3
2 1
1 9
8 7
6 5
0 0
0 9
9 9
9 9
1 1
1 0
0 0
0 0
)
g[rooa 9 '.= 48:s la E
~
>A.u* e$
m ksL r
t ~
t e
a~
s rb6oL J
e h
k uS T F ~~-
h.9@
=
60+
E 3*
00 1
5 0
+
- Lf l
l -
E 0
0 1
p 40+
E0 0
1 p
uo 3
r 0+
G E
0 8
3 0
2 1
7 r
e o
)
r f
V u
F 2
e g
0
(
C
+ F iF L
E C 0
L A
0 A C
1 C
E E
1 0+
E0
+
0 1
0 0+
E 0
.. j*
! j.
l
- 0 1
e
,+,
1 0
E0 I
0 g+ *4 1
, b 20 E
0 3
2 1
1 9
8 7
6 5 0 0
0 0
9 9
9 9
9 1
1 1
1 0
0 0
0 0
ey ir
vT.EanaO 2453
> 7_ w. cec L
r co6ow S
L_
U S.
- $Eo wc
=
= s 0
6 e
05 0
4 p
u e
o r
G
+
g 7
)
2 o
/
r 8
w o
(
e f
t r
n t
u 0 e n
e g
3 e
m i
F m
h h
ic r
c n
E i
rn E
0 2
0 Ig, 1
I 4
e 0i,I,,,
i #8 i
~
0 4
3 2
1 0
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 1
1 1
1 1
0 0
0 0
0
- I
In $n5:sO h 483
>7~ u. ee*
,if, l'
L CeD6ou ff S L e
k U S
- o=a wo
?
= -
+
~
j,;'
0 0
6 1
0 e_
0 4
1
+
0 0
2 1
0 0
0 p
1 u
+
+
o 9
r e
G
+.*
+
ru 7
g 2
0 X i
r F
0 /
o 8 H f
X H
I
+
006 e
$j
- 0 04 e.**
0 0
2 e,
+.,4
%g 8
a
?
l*
I O
4 3
2 1
0 9
8 7
6 5
4 0
0 0
0 0
9 9
9 9
9 9
1 1
1 1
i 0
0 0
0 0
0 e.
1
3 K :c Xa Q I.4g:s
.l' e
L
>7= N $c ff S L k'
U S 7e e8 2=a w ~
.l
=
6 0+
E
- ..h -
0 0
s
- +
1 k
5 0
s
+
E 0
0 1
+
+
4 0+
E 0
0 1
s 3
0 p
+
u E
o 0
r 0
G 1
7 0
2
)
V 1
r 2
e e
o 0
(
r f
+ F u
g F
E C 0 L L
i 0 A F
A C
1 C
E E
1 0+
E 0
0 1
0 0
+
+
E 4
0 0
1
+
+*
+
1 0
E 0
0
+
1
- 4). g+.
- 2 0-E 0
4 3
2 1
0 9
8 7
6 5
4 0 0
0 0
0 0
9 9
9 9
9 9
1 1
1 1
1 1
0 0
0 0
0 0
=y
g. e an2 O. ?.4$:s
- o
> 7= ea e3 L
r6* ooW S L tre U S_
x o
T!a wt E
l 06
~
0 5
0 4
p u
o r
G
- - +*
1
)o 1
r
/
o w
er
(
u f
tn g
tn e e 0 e i
3 F
e m
m hc h
w i
r c
n i
E rn E
0 2
31 I.'
+,
0 1
e*
e
- e*
e
+
,i
'. +
e l
l 1-h 0
4 3
2 1
0 9
8 7
6 5
4 0
0 0
0 0
9 9
9 9
9 9
1 1
1 1
1 0
0 0
0 0
0 tj
g xoX2 o p.45"
>5E
. oec L
ff S L 6* 8w U
S_
e k
5a O
=_
= u.
00
~
6
~
1
~
~
~
=
~
~
~
=
0
~
=
1
~
0
~
4
~~
~
~
=
~
~
=
=
0
~
0 2
=
1
=
=
=
000 p
4 1
u
=
o 9
r
=
2 G
1 6
=
er 1
u r
=
0 X g
o 0
I 8 H i
f F
+g
=
X
/
H
=
=
0
=
06
=
+
+
=
9 +
,*3
=
0
=
04
~
0 0
2
=
o# *
=
4, o T,o o
0 4
3 2
1 0
9 8
7 6
5 4
0 0
0 0
0 9
9 9
9 9
9 1
1 1
1 1
0 0
0 0
0 0
~
g5 xOon2 OL( o:
s
>7=
N*
e*
I L
ff S t r6* OOW ek U s m#o Wb
=
ji
=
N
+
M
~
~
~
2
~
1
~
~
00 p
g 0
1 u
o 3
G 1
+
e 6
ru 1
g r
F o
0 X i
0 /
~"
f 8 H
.{
X
/
~
H e
~
0
~"
06
~
e*
e*
~
~
~
0
~
0 4
n
~
~-
~
~
~=
~
n t
=
0 0
~
2
~=
9
+
' s 4
w g
- 0
+
~
3 a d *i 0
e 4
3 2
1 9
8 7
6 5
4 0
0 0
0 9
9 9
9 9
9 1
1 1
1 I
0 0
0 0
0 0
fe kI
g". k: n- @-
1
>?
. 3c rN T$
1 GVA L.
9 F
0 qo 8
)e 0
b o
O s
ac is O
h t
n o
i 7
sn 0 io 0
0 0t v
c 1 a r nf an h i 0
o Ts 6 n s
sf 0 o i
i t
slea ca Lm r
4 re F
Xh 1
n e
/ t I h o
r u
ig 5
I i
s g
rh 0 s i
o iF f o F
t ne l
t a
ou ib o
m t i r
cr t e
an 4 h r o 0 T Fc s nl a oc ii t
si s r ic Fn 3
o si 0
vt a mre d
ko m
C im 2
t
- e t
i 0
o oeL o g e N
i r u o o a r
(
el xvi a
uAF o
o-1 0
o O
o o
o^
c, u
0 4
3 2
1 9
8 7
6 5
i 0
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1
I'.FT2 o' oa8 p,
>3:" 3e*
C$68
- % w*
F I
e t
riu m
l i
9 iaL F
0 O
D g
o_
, 0 8
00 n
m 5
na h
O O
E 3
T s
7 U
0 se L
O O
dna 0
6 n 0
0 o
i 1
t n
n O
ca a
r h
F 5
T n
1 o
e r
5 r
e O
, 0 s
is u
O t
i g
ae F
iF r
O la G
m X
r
/
e I
4 h I
, 0 T ro O
f no i
n tc
, 3 a
0 r
F no t
i i
s m
s
,i 2
iF e0L 0
s g0 e O
v a1 r r
u n
e>l k
vXi e
a AHF
- a-
, 0 1
0 5
4 3
2 1
1 9
8 7
6 5
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
gCTT* nhoGS E
> 1 F 'oc*
r$68 T
w" 1
er u ;
ia ;,,
9 l
0 g3 F.,
A 4^E 6
8
, 0 a
0
, 7 0
05 na h
t r
6 n
e 0 o i
t t
a c
e a
r r
6 G
F n
1 X
o e
/
5 i
r s
u H
, 0 s
i g
r F
iF o
l f
a n
m o
r i
e c
, 4 h t
a 0 T r
F no iss 3
iF
, 0 sy m
t i
k m
i eL0 2
ge0 0
ar5 ru>
elviX a
AFH a
,1 0
a 0
5 4
3 2
9 8
7 6
5 1
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
3
g' :e co1 o ?.
c :
4z$
n
>7= ~ t-8e O
r6POOw mSa tO
- C 1
9 0
a e<..g^
7
+
8 0
e v.
+
s l
^
7 a
e.
0 c
t
+
i irC
)
\\
t e
l l
A 6
1 0
0 r
=
o f
7 n
n E
1 o
(
e i
n r
t o
u c
g a
+
5 it i
r 0 c a F
F r
F n
n o
io i
s s
s s
i i
4 F F
0 la la m
m re r
h e
T h
T
, 3 0
sv ffeK t
2 im i
0 eL g e a r r u el vi a
AF 1
0 j,
+
+
o ob< u
, ~.
L 0
5 4
3 2
0 0
0 0
0 9
9 9
9 9
8 7
5 1
0 0
0 0
0 1
1 1
1 1
?. _h,* a 48m
>7= 3. 3e i
?eebm
?e:o U 9,
0
+
, *+
+ +
8 0
e
$,+-,
no
+
ro
+
B
.V, h
7
, * +
0 h
t i
+
~
w s
+
)V la e
e c
i
, 6 1
t 0
i 0
r
=
C n
r E
o
(
8 f
no 1
n 5 i e
o t
0 c r
i a
u t
r g
ca
+
F iF r
no F
+
is n
s i
o 4 F i
0 l
s a
s i
+
m F
re l
h a
T m
+
r 3
e 0
hT sv ff
, 0 2
eK eg" ar "
evA*
I.
0 j'
0 9
8 7
6 5
5 4
3 2
1 1
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 l
1 1
1 1
I l
T': xoT* a 34m@
>7= " ~ ew r6cC8 mSaA 1
9 0
8 0
e t
i h
p a
r G
^
7 g
0 h
t i
w
)V s
t e
lac 6
1 0
i
, 0 t
=
r i
9 C
n 1
E r
(
e o
n r
f o
u 5 i g
n
, 0 c t
i o
a F
r i
c F
t a
no r
is F
s i
n 4 F o
0 la is m
s r
i e
F hT la m
30 re hT sv f
2 f
it m
0 e
i K
L e
e g r a u r l
e ia v FA 1
0 ji n-0 0
0 0
0 1
l 9
8 7
6 5
5 4
3 2
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
[E{ O=5'l
>1 g
C868
?is~
- S 1
9 0
"y
%x.
X x m 8
0 y
Mw X
x 4
X X g
)
7 n
gX
. 0 h
o i
t X
t i
w ar
)
s ne V
od M
l e
aio ct 6
1 am i
i r 0 0 t
=
r e n y
x Cdi f
of n
0 r
a E
oMr 2
(
v f
n a
e
)
o u
nd p r
5 i t
g oed X
0 c i t a
i t
an r
F c
F r a X
ar o n
o FBe X
i
- n ss
- r. n e i
4 F l
o o. h y
0 l
i r
s a
s( t i s e X
mr Fuy e
ol h
l o
T a np me rgr 3
e o e 0
h rt d a Tyw sl vI s e ff d
e u
2 K
c it e
0 l
n m
n O
e i
iL ly o
O
(
e n h g e i
t a
r r
f r
u e f
e O
A F WP ly v i t
a e
l a
r a
a o 1
P
. 0 O
oX e O
X X
X 0
g I
5 3
2 9
8 7
6 5
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
vE :oT
- o.
5'E :s c
=
>?= " ~ ee
~o
- dPC8
=5a A" I
9 0
,. i ^
g g
4 7
+
+
8 0
+
neg yx
+
. v O
7
. 0 h
t i
+
w
.+
s l
t
)
ac
\\
6 e i
. 0 t
1 ir 0
C
=<
ro n
f E
1
(
2 n
+
5 n o
0 e
io i
r t
tc u
c a
g a
r i
r F
F F
no n
. 0 4 i s o
s i
is F
s l
i a
F m
l r
a e
h m
3 T r
. 0 e
hT sv it f
m f
i 2
e eL 0
K g e a r r u el via AF 1
0 0
5 4
3 2
1 9
8 7
6 5
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
g". xO~M:s n 3" 8m
>7=
t-u $c r6 OOw c5a 4M 1
9 0
p.
- t. o4
+
e,
+
e 8
0 ss e
ln ia
+*+
7 t
0 S
0 h
+
t i
w
+
)V t
e s
la 6
1 c
0 0 i
=
+
t ir n
2 C
E 2
(
r n
e o
u 5 io r
f t
0 c g
n
+
a i
o F
r t
+
F i
c n
ar
+
io s
F s
i 4 F n
0 l
o a
s m
i s
r i
e F
hT la 3
m 0
re hT sv f
2 f
it
. 0 e
im K
eL g e a r r u el vi a AF 1
0
,4 4
n+
0 5
4 3
2 1
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 1
1 1
1 1
0 0
0 0
0
g'.:c T:1 a m 5'E" o
>7= F 8c r6?o8 T5o A*
I 9
, 0 p
,rt g
og
+
i^4 a
,e,
+
+
e 8
0
- %;m e*
e n
or I
. v h
,e+
7
, +
t 0
i w
+
s s+
l
)
a V
c e
i 6
1 t
i
, 0 r
0
=
C r
3 n
o E
2 f
(
e n
r n
o u
o 5 i i
t g
t 0 c iF c
ar ar
+
F F
no is n
s 0
l s
o i
4 F i
s a
i F
m r
l e
a h
m T
r e
, 3 0
hT sv f
it f
im e
eL 2
K g e 0
a r r u el vi a AF 1
i 0
,4 4
y:.
u+
0 5
4 3
2 1
y 9
8 7
6 5
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
g"' MOX2 n65',$n
>E G. g*
-th.8w mh N 1
9 0
- +
8 m
0 e
u n
i m
u 7
lA 0
h t
i w
)
\\
s t
e la 6
1 0
c 0
i
=
t i
4 r n
2C E
(
er n
r o
uf o
g 5
it in 0 c F o ra t
F i
c n
a o
r is F
is 4 F n
, 0 l
o a
m i
s r
s e
iF hT l
a 3
m 0
r e
h T
it m
s i
v eL 2
f g e 0
f a r e
r u el K
vi a AF 1
0 0
5 4
3 2
1 j
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 0
0 0
0 0
1 1
1 1
1
K' ;c Ta n45', ea o
d=.
F g*
7$,O8 2%*A 1
9
, 0
+
+ +.
+ +
mr
+
8 0
o F
%,+
+*
d
+
nu
+
+
7 o
p
^
7 4 +
, 0 m
{',
o
+
C
++
)
h V
t
^"
e i
w
, 0 6
1 0
s
=
la
+
+
c 5
n i
2 t E
i
(
e r n
uC r
o
+
5 i t
gr
+
0 c i
o a
F f r
F n
n o
io i
s tc s
i a
4 F r
0 l
F a
m n
r o
e h
is T
s iF 3
0 la mre hT 2
t s
i 0
m v
i f
eL f
g e e
a r K
r u el vi a AF
. 0 1
l 0
5 4
3 2
1 9
8 7
6 5
1 0
0 0
0 0
9 9
9 9
9 0
0 0
0 0
1 1
1 1
1
gU k:n'm5'Ug 1
>b E g*
Fb mk i
9 0
mr 8
0 o
e F
!. m.*
d nu op
^
m L
, 7 0
o C
)
h V
t t
e i
w 6
1 0
0 s
=
la c
5 n
i 2 t E
i
(
e r n
uC r
5 io t
gr 0 c i
o a
F f r
F n
n o
i o
i s
tc s
i a
4 F r
0 l
F a
m n
r o
e h
is T
s iF
,3 0
la mre hT 2
s it 0
v im f
eL f
g e e
a r K
r u el vi a AF 1
0 0
5 4
3 2
1 9
8 7
6 5
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
rE :ePr n n'.e. n', eo o
>'E G g*
r6* 8w mSa D i
9 0
m 8
0 ro F
la t
e 7
M 0
h t
i w
))e s
la 6
1 0
c 0
=
i t
i 6 r n
2C E
e
(
r r n
uo o
gf 5 i t
i 0 c F n a
o r
F i
tc n
a o
r is s
F i
4 F n
0 la o
m is r
s e
i h
F T
la 3
m 0
re hT sv 2
f 0
f eK eg *
+
ar "
e v
- A 1
0 oU l
0 l
5 4
3 2
1 9
8 7
6 5
0 0
0 0
0 9
9 9
9 9
1 I
1 1
l 0
0 0
0 0
E?:e r $ 6 5' Goa o
>? G-@*
C$ 8w
?% N I
9
, 0 4.
- /d W^
+
+
m 8
0 ro F
no i
t 7
u 0
lo S
h
)V t
i e
w s
, 6 1
0 l
0 a
=
c i
7 t n
E i
2 r
(
eC n
r o
ur 5
it go 0 c if a
F r
n F
o n
i o
t i
c ss a
i r
4 F F
. 0 la n
m o
r i
e s
h s
T iF 3
la 0
mre hT i
2 t
s m
v i
0 eL ff g e a r e
r u el K
i v a AF 1
0 0
5 4
3 2
1 g
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 0
0 0
0 0
1 1
1 1
1
g'.; X: n' mM' sE:s c
2
>3 G g
r$.8w t$ 8 I
9 0
+ e.
- +
+
, 8 0
5
+
y V
a s
- # +
7 sA g*
0
+
h t
+
i
)
w V
e s
6 1
l 0
a 0
c
=
i t
i n
8 r E
2C
(
e n
r r o
uo
, 0 c 5 i gf t
i a
F n r
o F
i t
n c
o a
is r
s i
F 4 F 0
l n
a o
+
m i
r s
e s
h iF T
l a
,3 0
mre hT s
2 v
t f
i 0
m f
i e
eL K
g e a r r u el vi a AF 1
0 0
1 5
4 3
2 9
8 7
6 5
1 1
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
>E :oF2 n 6 5',o:::
e
>7= F g*
r6*oS mea u*
J 1
9 0
p,
- 1#
4
+
+
+
0 8
1 0
l_ *+
y as sA 7
0 5
)V 0
e h
6 1
t i
, 0 0
w
=
s l
n 9 a 2 c E
(
i et n
ri o
u r 5
it gC 0 c i
a F r r
o
+
F f
no n
+
i o
ss i
i 4 F tc 0
l a
a r
m F
re n
h o
T iss 3
i 0
F la m
r i
t e
m h
i 2
T eL 0
g e s
a r v
r u el f
vi a f
AF eK 1
0 1
e
+
> U 0
0 0
0 0
1 i
9 8
7 6
5 5
4 3
2 0
9 9
9 9
9 0
0 0
0 0
1 1
1 1
l
s q
g :c Pr,: n' 4 $
>m2 i
g*
3
&6* 8w mSa 5 1
9
. 0 0
8 6
. 0 yass
+
A 7
0 0
)
1 Ve h
. 6 1
t 0
i 0
w
=
s 0l n
3 a E
(
c ei n
r t o
ui gr
. 5 i t
0 c iC a
F r
r F
o f
n n
ios i
is o
4 F tc 0
la ar mr F
e h
n T
o is 3
s 0
iF la mre 2
h 0
T s
y e
T g
f a
e re K
v
- 1 A
0 L
o 11 a
0 l>
F 5
4 3
2 1
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 0
0 0
0 0
1 1
1 1
1
c In ;og2
- E@m
>e E ". G $
Ce?C8
?=a M" I
9 0
8 0
06 y
a 7
ss 0
A h
t
)
i
\\
w e
s 6
1 l
0 a
0 c
=<
i t
1 i n
3 r E
eC
(
n ru r o
o 5
it gf 0 c i
a F n r
o F
i n
t a
io c
ss r
i F
,4 F
0 l
n a
o m
i r
s e
s h
T iF 3
la 0
m re h
T ll sv i
,2 t
m 0
ff i
eL e
g e K
a r r u el vi a
AF 1
0
+
4 1
0 5
4 3
2 1
9 8
7 5
0 0
0 0
0 9
9 9
9 1
0 0
0 0
0 1
1 1
1 1
K' :e g2 n b,4 h o
>7= F 3*
?e* oS
?
n
- ma t" 1
9 0
y r
t
+
+,
e m
+
oe 8
G 0
%fm..
e e
l iss i
F
. v g'
, +
7 la 0
c ir d
)
n V
t e
i ly 6
1 C
. 0 0
=
h t
2i n
3 w E(
e s n
rl o
ua 5 i t
gc 0 c i i t
a Fi r
r
+
F C
no r
is o
s f
i 4 F n
0 l
o a
m i
t r
c e
a h
r T
F n
. 3 0
o iss iF t
l i
a m
m i
2 e "c 0
g r
l e
s a -
h ie r u el r
T e via SAF
+
sy T
1 f
0 e
K 4
o's 0
5 4
3 2
1 g
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 0
0 0
0
'0 1
1 1
1 1
E ".c 8 n :s n 4 5', $ a
>E G g*
78 8w
- % U
?
1 9
, 0 y
- +
r te m
oe 8
G
, 0 e
l iss iF 7
la 0
c ire
)
h
\\
p e
S 6
1 0
h 0
t
=<
i 3 w n
3 s
E el
(
r a n
uc gi
,5 io t
t 0 c iFi a
r r
C F
r n
o o
f iss n
4 F i
o 0
l i
t a
c m
a r
r e
h F
T no 3
i 0
ss iF t
l i
a im m
eL 2
r ge 0
i e
s a r e r u h
el T
mvia sAF sy T
f 1
e l
K g
0 o
N 0
5 4
3 2
9 8
7 6
5 I
1 0
0 0
0 0
9 9
9 9
9 1
1 1
1 1
0 0
0 0
0
K : :::o T a n 6 5' m on
>t1 F g*
Ie* C8
?na M" 1
9 0
yr te m
8 0
oe G
e l
is
+
s iF
. 7 0
d io b
)
u
\\
C e
h
. 6 1
0 0 t
=
i w
6 n
3 s E
l
(
e a r c n
ui o
gi
+
5 i t
t 0 c ir a
F C r
F r
n o
o f
is n
is o
4 F i
0 l
t a
c m
a r
r e
F hT no 3
is 0
s i
F t
l i
m a
i m
eL g e 2
r a r e
r u 0
el h
vi a T
AF sv ff 1
eK 0
oU 0
5 4
3 2
1 1
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 0
0 0
0 0
1 1
1 1
1
7
.c g:1 n' m5' m8m
- o
>7:.F g*
r.e* o8 muoo u*
o I
9 0
+*
, 0 8
sy arr
+
.-v.
A 7
0
{
h t
i w
)
s V
la e
c 6
1 i
, 0 0 t
i o
r
+
C 5
n 3 r E
o
(
ef n
r o
un 5 i
+
, 0 c t
go i i a
Ft r
c F
a r
no F
iss n
i o
4 F i
0 l
s a
s m
i r
F e
h la T
m 3
r
, 0 e
h t
T i
m s
i v
eL f
g e f
a r r u 2
e el 0
vi K
a AF 1
0
>>P 0
4 4 1 1
9 8
7 6
5 5
4 3
2 1
0 0
0 0
0 9
9 9
9 9
1 1
1 1
l 0
0 0
0 0
s.
Kn ;c c.2 O' 5' p,S 8
>3 G GS C$.8w F% U 1
9 0
n
- W,
jg o
i t
+
a r
+
u
+
g 8
if 0
noC t
in U
7 0
e lg n
)V iS e
a 6
1 0
h 0
=
t i
6 w n
3 E
s el
(
ru a n
gi 5 io c
t it 0 c Fi a
r r
C F
r no o
f iss n
4 F i
o 0
l i
a tc m
a re r
h F
T n
3 o
0 iss i
t F
i m
l i
a eL m
g e 2
a r r
r u
, 0 el e
vi a
h AF T
+
+
s v
f 1
f
, 0 e
K
+
o 1,
+
0 5
4 3
2 1
l 9
8 7
6 5
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 I
1 1
1
g".n X2 n45' m o=
>3 3i g*
rN 8w m$ E I
9 0
+
8 g
0 6e l
s s
7 a
% +
0 M
h
+
t
)V i
w e
s 6
1 l
0 a
0 c
=<
i t
7i n
r E
3eC
(
n r r o
u 5 i o
t gf 0 c i
a F n r
o F
i n
t o
ca iss r
i F
4 F 0
la no mr is e
s hT iF 3
la 0
m t
r i
e m
h i
eL T
g e a r s
r u 2
v el vi 0
a f
AF feK 1
0 0
5 4
3 2
9 8
7 6
5 1
1 0
0 0
0 0
9 9
9 9
9 1
1 1
1 1
0 0
0 0
0
I.:eCW2 n6E' Gg
>7= P 3*
Ce* o8 2na u*
1 9
0 4g
[#
- +g^ +
g
+
6e 8
+
I 0
+
+
s sa
+
M 7
0 g
+
5e
)
l
+
V
^
e h
i
, 6 1
t 0
0 w
=
+
s 8
n la E
3 c
(
ei n
t r
o i
ur 5
i t
giC 0 c a F r r
F o
f no n
is o
s i
i 4 F t
. 0 c
l a
a
+
m rF re n
h o
it T
m i
i 3
s eL s
i g e 0
F a r r u el l
vi a a
AF mre h
2 0
T
+
sv ffeK 1
0 o4>
l>
+
l<
o 0
0 0
0 0
0 9
9 9
9 5
4 3
2 9
8 7
5 1
j 1
1 1
1 1
0 0
0 0
0
=
I' xPsna n' m5',E:s
>l=G. 3*
rb* 8w
=E$a $
1 9
0
+
8 g
, 0 m
5 e
I ssa
+
7 0
M h
t i
)
w Ve s
6 1
la 0 0 c
=
i t
i 9 r n
3C E
(
e r r no uo 5
gf 0 i tc iF n ar o
+
F i
t n
c o
a
+
is r
is F
4 F 0
n a
o m
i r
s e
s h
iF T
la 3
0 mr t
e i
h m
T i
- L E e s
r v
u 2
- l f
0 f
ia e
F K
+
4
+
1 0
o U
0 5
4 3
2 9
8 7
6 5
0 0
0 0
1 y
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
Kn:cPrna o' y,48:s
>7=G r6 8w m$a3 i
9 0
cc
+
8 5
0 e
+ -,
l
>e mu
+
7 l
0 oV h
)V t
i e
w 6
1 0
s 0
la
=<
c 0i n
t 4i E
r
(
erC n
u o
gr 5
it 0 c o
iFf ar F
n
+
o n
i o
t
+
is c
s a
i 4 F r
F i
0 l
t a
m m
n i
r o
eL e
i g e h
s a r s
r u T
i el F
vi a 3
AF l
0 W;
a mre hT 2
s 0
v i
f e
K
+
1 0
+
o l
0 l
9 8
7 6
5 5
4 3
1 3
1 0
0 0
0 0
9 9
9 9
9 1
1 1
1 1
0 0
0 0
0
1 g' :e8n: n' mgGg 1
>7=C g*
r6* 8w m5a O I
9
, 0
+ * * ^+
c
^
c 5
e 1
8
+
0 e
e m
u l
o V
, 0 7
c c
)
4
+
e V
t e
l 6
1 0
h 0
t
=
i
+
w n
1 4l E
s
(
e a n
r c o
ui 5 i t
t gi 0 c i r a
F C Fr r
n o
f ioss n
i o
4 F i
0 l
t a
c m
a r
r e
F hT no 3
i
, 0 ss iF t
l i
m a
i m
eL r
g e 2
a r e
r u 0
h, el vi a T
AF sv ff 1
eK 0
l o
lo 0
5 4
3 2
1 1
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 l
1 1
1 1
0 0
0 0
0 I
1' gn xosa n' e'dE#
f
>?=
1 g
0
- oPCS td' v!
- Ea 0" I
9 0
a
.r.
+ ++ /.'
i
- s d
+
e M
+
+
s 8
u 0
oenego
+
m 7
o 0
H e
t
)
in Ve ifn 6
1 0
I 0
=
h t
2i n
4 w E
(
t e s rl n n
u ae 5 io gcim 0 c t
i a
Ft t
r ia F
r e
Cr n
rT ios o
f is 4 F n
0 la o
m i
t r
c e
a h
r T
F n
3 0
o iss iF l
m it a
r im 2
0 L
+
e e
h r
+
u T
l
+
i a
s F
v f
1 f
, 0 e
+
K l>
o3 lt 0
3 7
1 i
9 8
7 0
0 0
9 9
9 0
0 0
1 1
1
vT. ;cPr 1 n4K',8m n:
>7=~
g*
-t6 8w m$a E I
9
, 0
+ +.
tn
- +
e m
+
e 8
0 t
a S.. **
er
+
T
+
l l
eC 7
'e
+
0 g
e c
+
i t
.+
t
)
a V
L
^'
e h
6 1
t 0 0 i
=
w
+
3 s n
4 E
la
(
e c n
ri o
ui
,5 i
t t
gr 0 c iFC ar r
F o
n f
o is n
is o
4 F i
t 0
l c
a a
s m
r re F
h n
T o
3 is 0
s F
it i
l im a
eL g e m
a r r u r
el e
vi
,2 a
h AF 0
T sv f
fe 1
K 0
0 5
4 3
2 1
1 9
8 7
6 5
0 0
0 0
0 9
9 9
9 9
0 0
0 0
0 1
1 1
1 1
K. c 7,~ n 6 j'o"
>t2 G. e
- Ce* 8w 2m* &
I 9
0 l
a
+..
ire ta M
e 8
0 r )n l_
+-
o i
t f
cf ea l
r fe a Rp 5
^
7
, 0 Z e, n
we
+.
l
)
oy V
Lh t
e t
ae 6
1 0
y 0
=
hl o tip n
4 w, E
4 r
(
s el e n
r u at o
c a 5
it gi w 0 c it a
Fi r
r n F
Ci a
n p
io rl o
s f,
i s
e 4 F nt
, 0 i
l o
a ih t p m
r c
j a
e a r h
i r
g m
F i
T r
eL 3
n g e e
ot a r 0
r u ia el s
vi a
iw AF s
F d l
e at a mr 0
2 r o eb h(
T sv ff 1
e 0
K 0
0 0
0 0
0 9
9 9
9 5
4 3
2 9
8 7
5 1
0 0
0 0
0 1
1 1
1 1
gn_%I*o 2o:E
>7= F Gee r6eb8 tso o*
E I
9 0
4+,
ro t
8 c
0 e
lf e
R l
a 7
te 0
M a)
)
h m V
t u
e iwi 6
1 n
0 s a 0
l r
=
a u c
n il E
5 tia
(
4 r r n
u o
e Ct r
5 it u
ra 0 c g
on a
f r
iF d
F n n n
o o
ia is t
cl s
i a e 4 F rk 0
l c
Fi a
m n n r
e ol h
i e
T ss e i t 3
F (s i
. 0 t
l im a
eL m
g e a r r
r u e
el vi h
a AF T
2 Wi 0
sv ffeK 1
0
,<0 t>
I<
0 1
9 8
7 6
5 5
4 3
2 1
0 0
0 0
0 9
9 9
9 9
1 1
1 1
1 0
0 0
0 0
lb K'.;cProa n6WUo"
>3 3i g*
C$.8m s' % O I
, 9 0
+
E+
4 n
o 4
i tc 8
e 0
lfeR tnac 7
i 0
f ing iS V) o e
N
, 6 1
0 0 h
=
t i
w n
6 E
4 s
(
l n
e a o
r c ui 5 i t
t 0 c gi a
ir r
F C F
n ro io s
f s
i n
4 F o
it
, 0 m
la i
i m
eL tc g e r
a a r e
r r u h
el F
vi T
a AF n
3 o
i 0
ss i
l F
la mr 2
e 0
hT sv ffe 1
K
, 0 4
0 5
4 3
2 g
9 8
7 6
5 1
0 0
1 1
1 0
0 0
9 9
9 9
9 0
0 0
0 0
1 1
g".:e 8a$' TE" e
>E i @*'
3 rh.8w mh E 1
9 0
o
+
+ +,
i ta R
D 8
O 0
0 leu
+
F h
+
/
c
^ 9 +
7 t
0 i
P 4
+
l le
+
c
+
)V e
e c
i t
f I.
t O J a
=
L
+
n 7 h t 2 E
4i
(
e wd n
o r
n 5 i u s a
t 0 c l
gaI a
iF c r
n F
i tie n
r e o
Cw iss rt i
e 4 F o
it m
0 fB i
m la n
eL
+
r o
g e a r e
i r u h
t el c
vi T
a a
AF r
F 3
0 n
o i
j, s
s iF 2
la 0
mre hT sv 1
0 ffeK 0
8 7
6 5
5 4
3 2
1 1
0 0
0 0
0 9
9 9
9 1
1 1
1 1
0 0
0 0
0
r v.;cPo$6EE"
>b G 5*
r 8m h $
1 9
0 o
i taR D
9 F
8 O
A
+
0 leu F
/
+
h i
7 c
t i
0 P
l l
e c
)
e
)
c e
i 6
1 t
t 0 0 a
=
L n
8 h t 3 E
4i
(
e wd n
o r
n 5 i us a
, 0 c t
l ga2
+
a iF c r
F in
+
tie n
r e o
Cw iss rt o
it i
e 4 F m
fB i
0 la n
eL m
o g e r
a r e
i r u h
t el c
vi T
a a
AF rF
, 3 0
no i
ss iF 2
la 0
mre hT s
1 v
. 0 ffeK 0
5 4
3 2
1 9
8 7
6 5
0 0
0 0
0 9
9 9
9 9
1 1
1 1
1 0
0 0
0 0
1 g" x8na n 238
>7=m 3c
-t6* 8w mSa 8 I
9 0
++
3 o
+
i taR D
O 7
l 0
eu F
/hc t
i 1
P 6 \\
l 0 e le 1
c 0
e
=
c i
t 9
t n
4 a5 E
Ld
(
5 n e
r h n 0
ito u
t a
i c
g w4 a
i r
F n
F s
l e
n ae
+
o icw iss t t 4 i ie it F
rB l
C C
m a
i.
m r
e1 r
o g e e
f a r r u h
n el T
via o
AF i
t fi 3
c 0
ar F
no iss iF 2
0 la mre hT s
1 v
0 ffeK 0
5 4
3 2
1 i
9 8
7 6
5 0
0 0
0 0
9 9
9 9
9 1
1 1
1 1
0 0
0 0
0