Molten Salt Tank - Materials 3

[Smith55]

Hi all,

We are at the feasibility stage of a project requiring a site built molten salt storage tank which has a maximum operating temperature of 435 Deg. C. This would be designed based on API 650 with some additional requirements considering the high temperature. My question is regarding material selection. As per my understanding P-No 1 Carbon Steels (E.g. SA516 Gr. 70) would be considered suitable for long term operation up to a maximum temperature of 425 Deg. C as specified in ASME codes, the temperature above which graphitization effect becomes significant.

Due to the challenges of carrying out PWHT on site are required for higher temperature steels such as C-0.5Mo, vendors we have spoken to propose moving to stainless steel (SS316 or SS347H) which would not require PWHT. This adds a significant cost increase when compared to a carbon steel tank.

My question is whether there are any code provisions or other materials or further assessment methods which could be employed whereby Carbon Steel e.g. SA516 Gr. 70 could be considered acceptable to a maximum operating temperature of 435 Deg. C. The project is in Europe and so we may also specify European codes. I think the answer is to proceed with stainless steel at higher expense but just wanted to check-in for any advice on this.

Any suggestions appreciated.

 

[MJCronin]

Let me offer a few comments about this increasingly popular storage tank design ... Flat bottomed, similar to API-650 configuration, but operating at a much higher temperature.... I assume that your system will have two tanks; a "cold tank", where carbon steel construction will be acceptable and a "hot tank" where a more expensive alloy will be required

The following comments are my opinion only ...

Your particular tank design question has been asked before on eng-tips ... But, the big problem is that there is no existing, acredited and established API, EN or EURO code that deals with the thermal expansion problems inherenent with these "Hot flat-bottomed tanks"

First, API-650/620 states does not apply to this high temperature range. There are inherent thermal expansion problems when a flat-bottomed tank undergos extreme heating and is "forced to slide" against some kind of flat foundation. Very high bending stresses are generated at the shell-to-bottom juncture... How do you intend to let this tank expand? ... Is there a proven method, or research paper that is telling you how to do this ?

Second, You simply cannot pick and chose any rules from any standard that you want. You cannot chose material stress limits from one code (ASME-VII) and use design rules from another code

Third, I believe that you will have to use a comprehensive, "BIG GUN" methodology, such as ANSYS to evaluate your shell-to-bottom joint design

Fourth, IMHO you would be begging for disaster if you use a common Carbon Steel material(SA516 Gr. 70) at 435C (815F).. clearly an established range where slow graphitization and ultimate failure WILL OCCUR !!! ... The lawyers, will pounce on this in the courtroom.... You will be called to testify many years after this project is in your past

Fifth, I believe that you should consider the use of the slightly more expensive SA 240-304L/316L stainless material for your "hot tank"... Sooner or later, the operators will "accidently" run the system above 435C ... Your boss will then prance over to your desk someday and ask you (on the spot) if you can approve a recent excursion to 500 - 600C... is it OK ?... Can you give the owners a response by 2:00 pm this afternoon ? .... You MBA boss will scream at your suggestion of the more expensive stainless steel material ... Without an engineering degree, it is all that he can do .... 304L/316L is kind of the "bottom shelf" in SS resistance to corrosion ...You may be pushed into a more corrosion resistant SS by industry experience

Sixth, What design methodology are you using to guarantee uniform expansion of the flat-bottom against the foundation ?... Have you found any EXISTING hot tanks that have been operating for years without problems ? ... Please share your methodologies and research

Seventh, ... Out of curiosity, what kind of weld detail do you propose for your "shell-to-bottom" hot tank joint ? (API-650 requires a double fillet weld, inside and out) ... This will be the most highly stressed area of your hot tank AND most subject to failure..... I suggest a full penetration joint with fillet back up inside and outside ... Have you done any document research about the design of this "hot joint" ?

Tell us more details about the design ... What are you proposing for nozzle design ?.... How will the hot tank be anchored ?....How is the tank vented ?... What are the tank dimensions ?

MJCronin
Sr. Process Engineer

 

[MJCronin]

This NREL "Lessons Learned Report" link may be useful: (pages 211-216)

https://www.nrel.gov/docs/fy20osti/75763.pdf

The tanks discussed selected SA240-Type 304 stainless plate materials with a special internal salt distribution system AND an elaborate and special FOAMGLASS foundation ...

Your BH MBA boss will never permit such a complex foundation design and ask you to come up with "something else"..

Another PhD paper discusses the problems with the design methodologies (page 67) and distrubution of hot incoming salt

https://www.tdx.cat/bitstream/handl...d=709D0F14009D1CE7AFB6ADAEC74D6B1A?sequence=2

My opinion only ....

MJCronin
Sr. Process Engineer

 

[Smith55]

MJCronin thank you for your feedback which all makes sense.

Yes - "cold" tank in carbon steel and "hot" tank in more expensive alloy.

Yes, agree that there are no existing and accredited codes. The experience in this area is in the CSP industry where there have been lots of lessons learned over the last few tanks as these storage tanks have been scaled up. The parabolic trough type CSP technology is the most mature which operates below 400 Deg. C and hence use Carbon Steel tanks for both hot and cold tanks. Solar tower type CSP operates at a higher temperature of approx 560 Deg. C and I understand that SS347H or SS316 is most commonly used for the hot tank at this higher temperature range. In terms of sizes, Carbon steel tanks up to about 50m diameter with shell height around 15m are in operation at several sites. Hot tanks in SS347H have also been built in this size range but have apparently had a number of problems at these larger sizes in the last few years.

Carbon Steel at 435C - note your comments on this regarding graphitization which makes sense an aligns with my reading on this. Can you suggest any good reference books / papers which provides time (years of service) Vs temperature graphitization charts, or further information on this which would support my report?

SA 240-304L/316L - I would like to find some references of this material being used for this application, since the references I have indicate SS347H has been used more commonly in the CSP industry for the larger tanks (I believe due to higher corrosion resistance at the 560 Deg. C range).

To mitigate the risks we will be retaining a competent engineering consultant to undertaken the detailed engineering and a vendor with strong track records in the CSP industry, and we would be looking to employ a size, design and material combination which has been proven in industry.

Regarding your comments on uniform expansion, shell to bottom joint and other queries - note your comments and I don't have the answers to your questions at the moment - something I will pay close attention to and obtain further advice.

Thanks for the research paper links.

 

[HTURKAK]

I have past experience for hot tanks but for hot bitumen, resid etc with a design temp around .200 Degr.
However , IMHO , molten salt storage tanks which has an operating temperature of 435 Deg. C. and probably future 550 Deg. C is totally different story.


In general i agree with Mr CRONIN and a pinkk star for his respond . But still i have the opinion that the use of SA516 Gr. 70 together with internal refractory brick and incoloy lining could be an option ...and IMO, it is worth to dig more this subject.

I have written ( INVESTIGATION OF THERMAL STORAGE ISSUES ) and search the WEB ..One of the outcomes, an old but gold document ,

https://digital.library.unt.edu/ark:/67531/metadc1403660/m2/1/high_res_d/10189397.pdf

Pls have a look in detail ( Section 4 Thermal Storage System Hot Salt Tank Designs and Cost Estimates ) this document and pls keep us involved in your final decision on these important tanks...








If you put garbage in a computer nothing comes out but garbage. But this garbage, having passed through a very expensive machine, is somehow ennobled and none dare criticize it. ( ANONYMOUS )

 

[Smith55]

Thank you HTURKAK for the input and for the document link. The design option in this document as proposed by S, N. Technigaz looks very similar to a membrane type LNG tank? I'm not aware that this design option has been used by any of the commercial CSP projects for storage of nitrate salts.

 

[MJCronin]

Smith55 ...

Please keep us in the loop for your final decision on these tanks and their design features ...

Please share what you can ....

This is a very imporatant and interesting thread, IMHO ...

Best Regards

Michael J Cronin, PE

MJCronin
Sr. Process Engineer

 

[HTURKAK]

- True.. similar approach.. S.N. Technigaz developed stainless steel membrane system for LNG tanks with two sets of corrugations running at right angles to minimize the thermal contraction stresses.

- Similar corrugations shall be studied for molten salt hot tanks if stainless steel membrane system with SA516 Gr. 70 shell is selected..

- The preliminary elevation ( Figure 4-3, mit 600 mm RC mat ftg) does not include foundation cooling . The inner temp. 435 Deg. C. and future 550 Deg., will cause temperature build up under the foundation . Found. cooling , ( raising the mat on short columns etc) shall be considered ,
- The thickness of refractory brick und external insulation shall be studied carefully to get stabile temperature gradient and max steel temp . 350 Deg. C. together mit minimum heat loss..

My opinion..






If you put garbage in a computer nothing comes out but garbage. But this garbage, having passed through a very expensive machine, is somehow ennobled and none dare criticize it. ( ANONYMOUS )