ML20140D233: Difference between revisions

From kanterella
Jump to navigation Jump to search
(StriderTol Bot insert)
 
(StriderTol Bot change)
 
Line 164: Line 164:
i the heat removal from the pool is conservatively minimized by neglecting heat loss through convection and evaporation from the surface of the pool to the fuel building atmosphere and by neglecting heat conduction from the water through the walls of the pool.        By maximizing the heat addition and minimizing the heat removal, the spent fuel pool water temperature is conservatively maximized.          In spite of this conservatism, it may l
i the heat removal from the pool is conservatively minimized by neglecting heat loss through convection and evaporation from the surface of the pool to the fuel building atmosphere and by neglecting heat conduction from the water through the walls of the pool.        By maximizing the heat addition and minimizing the heat removal, the spent fuel pool water temperature is conservatively maximized.          In spite of this conservatism, it may l
be seen that the temperature of the pool is below boiling and the l
be seen that the temperature of the pool is below boiling and the l
: 3)        In the attachment to a letter dated October 5, 1984, from J.W. Williams, Jr. (FPL) to Steven A. Varga (NRC), p. 4, it was stated that the evaporation rate / required makeup was 37 gpm. The 37 gpm refers to the bolloff rate in the event of
: 3)        In the attachment to a {{letter dated|date=October 5, 1984|text=letter dated October 5, 1984}}, from J.W. Williams, Jr. (FPL) to Steven A. Varga (NRC), p. 4, it was stated that the evaporation rate / required makeup was 37 gpm. The 37 gpm refers to the bolloff rate in the event of
!                    boiling, which is discussed later in the affidavit.
!                    boiling, which is discussed later in the affidavit.



Latest revision as of 17:33, 12 December 2021

Affidavit of DC Patton Addressing Contentions 6 & 8.Util Committed to Upgrade Spent Fuel Pool Cooling Loop to Ensure Adequate Cooling of Spent Fuel Pool in Event of SSE
ML20140D233
Person / Time
Site: Turkey Point  NextEra Energy icon.png
Issue date: 01/22/1986
From: Patton D
BECHTEL GROUP, INC., FLORIDA POWER & LIGHT CO.
To:
Shared Package
ML20140C819 List:
References
OLA-2, NUDOCS 8601290230
Download: ML20140D233 (19)


Text

- -. -_

ll

9. t J N.

N' 'N

  • 01 N
)' \

5 be p._

UNITED STATES OF AMERICA F- O '

NUCLEAR REGULATORY COMMISSION

\

?,  :

BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 's'x /

V

'Q.

)

In the Matter of ) Docket Nos. 00-250-OLA-2

) 50-251-OLA-2

' FLORIDA POWER AND LIGHT COMPANY )

)

(Turkey Point Nuclear Generating )

Units 3 & 4) ) (Spent Fuel Pool Expansion)

)

AFFIDAVIT OF DANIEL C. PATTON

_ ON CONTENTION NOS. 6 AND 8

1. My name is Daniel C. Patton. I am employed by Bechtel Power Corporation, Eastern Power Division, as a Senior Engineer in the Thermal-Hydraulics Group of the Nuclear Engineering Staff. In this position, I was involved in the thermodynamic and heat transfer calculations performed in support of the expansion of spent fuel storage capacity at Turkey Point Units 3 and 4. A summary of my professional qualifications and experience is attached as Exhibit A and is incorporated herein by reference.
2. The purpose of my affidavit is to address Contentions 6 and 8. Contentions 6 and 8 end the bases for the i Contentions are as follows:

i 8601290230 860123 PDR ADOCKOSOoggO O

Contention 6 The Licensee and Staff have not adequately considered or analyzed materials deterioration or failure in materials integrity resulting from the increased generation and heat and radioactivity, as a result of increased capacity and long term '

storage, in the spent fuel pool.

Bases for Contention 6 The spent fuel facility at Turkey Point was originally designed to store a lesser amount of fuel for a short period of time. Some of the problems that have not been analyzed properly are:

(a) deterioration of fuel cladding as a result of increased exposure and decay heat and radiation '

levels during extended periods of pool storage.

(b) loss of materials integrity of storage rack and pool liner as a result of exposure to higher levels of radiation over longer periods.

(c) deterioration of concrete pool structure as a result of exposure to increased heat over extended periods of time.

Contention 8 That the high density design of the fuel racks will cause higher heat loads and increase in water temperature which could cause a loss-of-cooling accident in the spent fuel pool, which could in turn cause a major release of radioactivity to the environment. And, that the decrease in the time that it takes the spent fuel to reach its boiling point in such an accident, both increases the probability of accidents previously evaluated and increase [ sic) the chances accidents not previously evaluated.

Bases for Contention 8 a) The NRC has stated in numerous documents that the water in spent fuel pools should normally be kept below 122 degrees F. The present temperature of the water at Turkey Point is estimated to be 127 degrees F. After the reracking, the temperature of the water could rise to 141 degrees on a normal

i basis, and could reach 180 degrees F. with a full core load added. In addition, the time for the spent fuel boiling point to be reached in a loss 1

of cooling accident will go from 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> to 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. Four hours is clearly not enough time to take action to prevent a major accident in the spent fuel pool from occurring. Thus, the increase in heat and radioactivity resulting from increases [ sic] density will result in an increase in the probability of a major spent fuel pool meltdown occurring.

b) There is also the possibility that a delay in the i

make up emergency water, could cause the zirconium

) cladding on the fuel rods to heat up to such higher temperatures that any attempt at later cooling by injecting water back into the pool could has'.en the heat up, because water reacts chemi.cally with heated zirconium to produce heat and possible explosions. Thus, the zirconium cladding could catch on fire, especially in a high density design,and create an accident not

previously evaluated.

Specifically, the purpose of my affidavit is to address the generation of heat and the cooling of the spent fuel as a result i

of the increase in storage capacity of the Turkey Point spent fuel pools. With respect to Contention 6, the Affidavit of Rebecca K. Carr on Contention No. 6, the Affidavit of Eugene W.

Thomas on Contention No. 6, and the Affidavit of Dr. Gerald R.

, Kilp on Contention No. 6 discuss the matters related to materials degradation.

Decay Heat Generation 1

3. As a result of the fission process in the reactor, radioactive fission products exist in the fuel assemblies during I

and af ter reactor operation. It is the decay of these fission products which produces heat in the spent fuel. The amount of

the heat given off by the spent fuel is a function of the time the fuel was in the operating reactor (burnup) and the time after the cessation of operation or shutdown (decay time). As shown in Figure 1, the amount of decay heat generated by spent fuel decreases with time following shutdown (the decrease is exponential in nature).

4

4. During a refueling, a portion of the reactor core is transferred from the reactor vessel to the spent fuel pool and is replaced by new fuel in the reactor. The spent fuel pool 4

serves as a storage facility for the spent fuel, providing radiation shielding and decay heat removal. The shielding and l heat removal functions of the spent fuel pool are assured by maintaining an adequate water level in the pool and through the

operation of the spent fuel pool cooling system.

Soent Fuel Pool Coolina System

5. The spent fuel pool cooling system for each unit l

1 consists of a pump, heat exchanger, filter, demineralizer, piping and associated valves and instrumentation. Redundancy of this i equipment is not required because of the large heat capacity of the pool and its corresponding slow heat-up rate. Nonetheless, a 100-percent-capacity spare pump which is permanently piped into the spent fuel pool cooling system has been installed. This pump

, is capable of operating in place of the originally installed

.i

?

-. . .. - . . . _ . _ _ _ _ . . . , _ _ , _ , , , _ , . _ . , . . . . ~ . . _ _ _ _ _ _ _ _ , _ . _ - _ . _ _ _ _ _ _ , _ . . _ , , _ , _ _ - . . _ . _ _ . _ _ _ . . _ . . _ ._.. ,_ , . . - . _ . , _ _

0 1

0 I 8 E

M I

T S I 6 V

T U

P )

T C 1 U #

eO r

S

(

uT gA E

i E M I

F H T D

E Z

l l

A I 4 M

R O

N 1 2

- - 0 9

0 8

0 7

0 g' 2sE $b3O IOwN342zOZ

,l!l lll, :ll ; i:i,i! :4I1 ,!i:i!: iil:i i < 1 :iii

. l crmp, but not in parallel with the original pump. Also, alternate connections are provided for connecting a temporary pump to the spent fuel pool loop.

6. The removal of heat from the spent fuel pool is i

I accomplished primarily through the spent fuel pool heat exchanger i

(a shell and U-tube type heat exchanger with the spent fuel pool water flowing through the tube side and the plant component cooling water flowing through the shell side). The heat removal

! rate through the heat exchanger is a function of the temperatures of the component cooling water and the spent fuel pool water.

For example, for a component cooling water temperature of 100UF and the design flow rates of 1.4 x 10 6 lbm/hr for the component i cooling water and 1.1 x 10 6 lbm/hr for the spent fuel pool water, I. the heat exchanger would remove approximately 8.0 x 10 6 Btu /hr when the spent fuel pool temperature is 120 0F and approximately 0

l 16.0 x 10 Btu /hr when the spent fuel pool temperature is 140 F.

7. It is expected that some water will be lost from the spent fuel pool by evaporation from the pool surface. The amount of this evaporation also depends on the pool temperature as discussed in paragraphs below. The normal makeup to the pool i

l 1s supplied from the demineralized water system or from the refueling water storage tank through the refueling water purification pump at a flow rate of up to 100 gpm. Alternate means of makeup also include temporary connections from the fire water sjatem or the primary water storage tank.

1

., l i

8. The existing cooling system piping and make-up supply lines for the spent fuel pools are not Seismic Category I and have not been designed to remain functional after a safe shutdown earthquake. In response to questions from the NRC
staff, FPL committed to upgrade the spent fuel pool cooling loop such that it will remain functional after a safe shutdown earthquake. 1/ This upgrade wil; be completed by the end of the second refueling outage after issuance of the amendments for the spent fuel pool expansion 2/ Until the upgrade is completed, the amount of fuel assemblies that will be stored will be less than the capacity of the pre-existing storage racks as 1 demonstrated in Table 1.

i f

l l

l i

l 1/ Letter dated October 29, 1984, from J.W. Williams, Jr. (FPL) j to Steven A. Varga (NRC).

2/ Safety Evaluation related to Amendment No. 111 to facility operating license No. DPR-31 and Amendment No. 105 to facility operating license No. DPR-41 for Turkey Point Unf.hs 2

3 and 4 (November 21, 1984), p. 15.

TABLE 1 Number of Assemblies To Be Stored in the Turkey Point Soent Fuel Pools Turkey Point Unit 3 Approx.

Total No.

Assemblies Approx. in Pool l Cycle from all l Startup Previous Cycle No. Date Cycles Comments 1 to 8 --

311 --

9 1/7/84 369 Amendment issued 11/21/84 10 7/17/85 425 --

- 11 3/7/87 481b/ Seismic Upgrade complete by end of Cycle 11 refueling outage 1/ The pre-existing storage racks for Unit 3 had a capacity of 621 assemblies.

Turkey Point Unit 4 1 to 8 --

287 --

9 05/16/83 323 --

10 6/1/84 387 Amendment issued 11/21/84 11 3/30/86 455 --

12 12/15/87 503b/ Seismic Upgrade complete by end of Cycle 12 refueling outage 2

1/ The pre-existing storage racks for Unit 4 had a capacity of 614 assemblies.

\

Normal Refuelino Offload i

9. For normal operations, the amount of fuel that is removed from the reactor vessel and placed in the spent fuel pool I

during refueling is between 1/3 and 1/2 of the reactor core. The analysis performed in support of the spent fuel storage facility expansion assumed that 1/3-core would be loaded into the spent f fuel pool for the first six refuelings, and, for the next 14 refuelings, 1/2-core would be loaded into the pool for a total of nine cores (6 x 1/3 & 14 x 1/2). The nine cores are nine assemblies more than the maximum capacity which the new high-

! density storage racks can accommodate.

l 10. As discussed above, the amount of decay heat given j off by each spent fuel assembly is a function of the burnup and the decay time. The decay heat was calculated in accordance with the Nuclear Regulatory Commission's (NRC's) Branch Technical i

Position ASB 9-2, " Residual Decay Energy for Light Water Reactors

for Long Term Cooling." The total heat load to the pool following refueling is also a function of the amount of fuel placed in the pool (offloaded). The maximum heat load for a j normal refueling offload is calculated to occur with 8-1/2 cores l

stored in the pool and 1/2-core offloaded at refueling. This results in a peak total heat generation rate of 17.9 x 106

! Btu /hr, which is calculated to raise the pool water temperature I to 143 F. Then, due to a decreasing heat generation rate and 1

due to the continued operation of the spent fuel pool cooling l

- l

\

I i

l -

i

_g_

system, the pool water temperature is calculated to decrease to 140 F in 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> after reaching the peak and to 130 F in 720 hours0.00833 days <br />0.2 hours <br />0.00119 weeks <br />2.7396e-4 months <br /> after reaching the peak. This is depicted graphically in Figure 2. At the peak temperature, the water loss from the pool through evaporation is calculated to be approximately 1.5 gallons

) per minute, which is well within the capacity of the makeup system. 3]

, 11. The calculation of the heat load to the pool is

] conservatively maximized by assuming that the entire 1/2-core offload is added to the pool exactly 150 hours0.00174 days <br />0.0417 hours <br />2.480159e-4 weeks <br />5.7075e-5 months <br /> after reactor shutdown. Thus, credit is taken for only 150 hours0.00174 days <br />0.0417 hours <br />2.480159e-4 weeks <br />5.7075e-5 months <br /> decay. In reality, fuel offload at Turkey Point typically does not begin until well after 150 hours0.00174 days <br />0.0417 hours <br />2.480159e-4 weeks <br />5.7075e-5 months <br /> after shutdown and extends over a l period of time such that much of the decay heat is given off l before the fuel is placed in the spent fuel pool. In addition, f

i the heat removal from the pool is conservatively minimized by neglecting heat loss through convection and evaporation from the surface of the pool to the fuel building atmosphere and by neglecting heat conduction from the water through the walls of the pool. By maximizing the heat addition and minimizing the heat removal, the spent fuel pool water temperature is conservatively maximized. In spite of this conservatism, it may l

be seen that the temperature of the pool is below boiling and the l

3) In the attachment to a letter dated October 5, 1984, from J.W. Williams, Jr. (FPL) to Steven A. Varga (NRC), p. 4, it was stated that the evaporation rate / required makeup was 37 gpm. The 37 gpm refers to the bolloff rate in the event of

! boiling, which is discussed later in the affidavit.

Flow. 2 SPENT FUEL POOL WATER TEMPERATURE VS TIME AFTER SHUTDOWN

INormal Refueling)

Tmax = 143 F I tag -

1 l

E w

a i E E

z g 130 -

e i 1 r

d r 2 .

~

!E E 120 -

j .

1 i

1 l

]

i i

110 I I I I ,

l '

1 0 200 400 800 800 , 1000

]

TIME AFTER SHUTDOWN IHRS)

i l

I mass loss through evaporation is well within the makeup capacity of the system. NRC Standard Review Plan Section 9.1.3, " Spent Fuel Pool Cooling and Cleanup System", paragraph III.l.d, states that the temperature of the pool should not exceed 140 0F and the liquid level in the pool should be maintained for the maximum normal heat load with normal cooling systems in operation.

Although the peak calculated pool water temperature slightly 0

exceeds 140 F (by 03 F), the calculation is conservative, as stated above, and the analysis shows the temperature decreases to 0

below 140 F within a relatively short time (72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />).

Consequently, the spent fuel will remain adequately cooled and covered with water at all times.

Full Core Offload

12. An analysis of a full core offload was also performed using assumptions similar to those in the analysis of the normal core offload. However, for the abnormal case of a full core offload, the analysis performed for the increase in spent fuel storage capacity assumed eight cores stored in the pool and one core offloaded. This results in a maximum heat load to the pool of 35.0 x 10 6 Btu /hr which causes the temperature of the pool water to increase to a peak value of 183 F. At this temperature, the heat removal rate of the spent fuel pool cooling system has increased to the point where it approximately equals the heat generation rate from the stored spent fuel. Thereafter, the decay of the relatively short-lived fission products in the

)

spent fuel causes the heat generation rate to decrease. At the l peak temperature of 183 0F, the rate of mass loss through evaporation from the surface of the pool is conservatively calculated to be 5.5 gallons per minute. 4/ Thus, it may be seen that the pool temperature is below the point of boiling, that the evaporation rate is well within the makeup capabilities of the system, and that the spent fuel pool cooling system is adequate to remove the heat generated from the stored spent fuel. Thus, there is assurance that the spent fuel will be cooled and covered with water at all times. SRP Section 9.1.3, paragraph III.l.d

! states that the temperature of the pool water should be kept below boiling and the liquid level in the pool should be maintained for the abnormal maximum heat load from a full core i

offload with normal systems in operation. The analysis performed

shows the cooling system to be in compliance with the SRP guidelines.

Loss of Soent Fyel Pool Coolina

13. General Design Criterion 61 of Appendix A to 10 CFR Part 50 sta.tes that fuel storage and other systems which contain radioactivity shall be designed "with a residual heat

~

removal capability having reliability and testability that reflects the importance to safety of decay heat and other j

g 4/ In the attachment to a letter dated Octeber 5, 1984, from i J.W. Williams, Jr. (FPL) to Steven A. Varga (NRC), p. 5, it

, was stated that the evaporation rate / required makeup was 72 gpm. The 72 gpm refers to the boiloff rate in the event of boiling, which is discussed later in this affidavit.

l I

_ _ . - . . . _ . ~ . - . . - . _ _ _ _ _ , - . _ . . , _ . . . _ _ _ . . . . _ .-_ . _ . - . _ _ _ _ - _ - - _ _ _ _ _ . . _ . _ _ _ . _ . . ,

residual heat removal." The NRC Staff has provided guidance for satisfying this requirement in Standard Review Plan Section 9.1.3. Paragraphs I.1 and III.1.b of Standard Review Plan Section 9.1.3 state that the spent fuel pool cooling system should be Seismic Category I or, in the alternative, the make-up system, the fuel pool building, and the ventilation and filtration system should be Seismic Category I and be able to withstand a single active failure.

14. As discussed above, the cooling system piping and make-up supply lines are not Seismic Category I. Accordingly, the NRC Staff requested FPL to perform an analysis of loss of cooling to the spent fuel pool, and the safety analysis report for the spent fuel pool expansions 5/ and FPL's responses to NRC questions 6/ provided the results of analyses of loss of cooling to the spent fuel pool. The analyses showed that, with all positions in the new storage racks full with assemblies, the pool would not begin to boil for a normal offload until a minimum of 7.6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> after loss of cooling and 1.6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> after loss of cooling for a full core offload, which, as stated above, is an
abnormal case.

5/ " Spent Fuel Storage Facility Modification Safety Analysis Report", Florida Power & Light Company, Turkey Point Units 3 i and 4 (March 14, 1984) pp. 3-12 to 3-13.

l 6/ Letter dated July 2, 1984, from J.W. Williams (FPL) to l l

Steven A. Varga (NRC), Attachment, p.l., letter dated '

October 5, 1984, from J.W. Williams, Jr. (FPL) to Steven A.

Varga (NRC), Attachment, pp. 4-5.

I l

15. In response to the NRC review of FPL's application N

for the spent fuel pool expansion, FPL has committed to upgrade its spent fuel pool cooling loops to ensure that they will remain functional after a safe shutdown earthquake. Therefore, after the seismic upgrade, the Turkey Point spent fuel pool cooling ,

system will comply with paragraphs I.1 and III.l.b of SRP Section 9.1.3 and will provide adequate cooling in the event of an earthquake. As discussed in paragraph 8 above, this upgrade will be completed by the end of the second refueling outage after issuance of the spent fuel pool expansion amendments, and prior to the time the cooling system is upgraded, the spent fuel pool will only contain the number of fuel assemblies authorized prior to the issuance of the Turkey Point spent fuel expansion amendments. The spent fuel pool expansion amendments will not result in an increase in the amount of cooling and makeup necessary for these assemblies.

16. Finally, it should be noted that zirconium-water interaction is not a concern with respect to storage of spent fuel assemblies at Turkey Point. Zirconium and water or steam can interact to produce heat; however, this reaction does not occur at temperatures below 1000 F. As demonstrated above, adequate cooling and make-up will be provided for the stored spent fuel assemblies at Turkey Point, resulting in temperatures far below those necessary for a zirconium-water reaction.

. - 14 _

CONCLUSION

17. For both a normal and full-core offload, the Turkey Point spent fuel pool cooling system and make-up system have adequate capacity to maintain the temperature of the spent fuel pool water below boiling and to keep the spent fuel covered with water, thereby ensuring adequate cooling for the spent fuel.
  • Additionally, FPL has committed to upgrade its spent fuel pool cooling loop to ensure adequate cooling of the spent fuel pool in the event of a safe shutdown earthquake.

I t

i 1

r

-- , - - . , - . - - - . --. . . -+-_, _ - -. - -

FURTHER AFFIANT SAYETH NOT 1

The foregoing is true and correct to the best of my l

\

4 knowledge, information and belief.  !

'e ~

Daniel C. Patton STATE OF MARYLAND )

COUNTY OF MONTGOMERY)

Subscribed and sworn to before me this Nk day o AA<L , 1906. My commission expires:

My Commission Expires July 1,1986 NOTARY PUBLIC

1 i

EXHIBIT A STATEMENT OF PROFESSIONAL QUALIFICATIONS OF DANIEL C. PATTON CURRENT POSITION Senior Engineer, Bechtel Power Corporation EDUCATION BSc, Nuclear Engineering, North Carolina State University, May 1977 MS, Mechanical Engineering, University of Maryland, May 1983

SUMMARY

OF EXPERIENCE WITH BECHTEL 4-1/4 Years Group leader and senior engineer, Nuclear Staff, July 1980 - Present 3 Years Engineer, Nuclear Staff and nuclear power plants, June 1977 - 1980 1-1/2 Years Engineering assistant, Nuclear Staff,

, (cumulative) alternating semesters between August 1973 -

1 August 1976.

a EXPERIENCE WITH BECHTEL 1

Mr. Patton is currently serving as group leader for the Nuclear Staff thermal hydraulics group. He is responsible for supervision of thermal hydraulic containment and subcompartment analyses, heat transfer analyses, combustible gas analyses, and approval of thermal hydraulics calculations.

Prior to the above, as a senior engineer with the thermal hydraulics groups, Mr. Patton performed thermal hydraulic containment and subcompartment analyses for high energy line breaks, combustible gas analyses, heat transfer and equipment qualification analyses, and coordinated the turnover and evaluation of thermal hydraulic calculations for the Houston Lighting & Power Company's two 1250 MW PWR unit South Texas Power Plant.

l Previously, Mr. Patton was engaged in examination and  !

evaluation of decontamination equipment and techniques for the Metropolitan Edison Company's Three Mile Island Unit 2 recovery project, as well as preparation of procedures for decontamination of the containment. He also participated in I.E.Bulletin 79-14 walkdown inspection of safety-related and seismic Category I piping and verification of as-built drawings at Northeast Nuclear Energy Company's Millstone

! Nuclear Power Station 870 MW PWR Unit 2.

In addition, Mr. Patton was involved in review of drawing revisions and design modifications for the reactor coolant system, residual heat removal system, accumulator safety injection system and borated water storage tank for the multi-unit SNUPPS Project 1150 MW PWR units. As the responsible engineer for certain mechanical components, his work included coordination with equipment vendors, site engineers, and other disciplines.

Former staff assignments included special studies for the Three Mile Island containment decontamination and recovery project, as well as secondary system modifications for Florida Power & Light Company's Turkey Point Power Plant 760 MW PWR Units 3 and 4. Mr. Patton was also engaged in computer code maintenance which involved implementation of modifications, documentation and verification of computer codes used in containment and subcompartment thermal hydraulic analyses.

Earlier, as a co-op student on the Nuclear Staff, Mr. Patton gained experience in radwaste, shielding and thermal hydraulics.

PROFESSIONAL AFFILIATIONS American Nuclear Society

)

. _ . --_ .- _. _ . _ - -