9 replies to this topic

[damik]

Dear forum members,

I am working on a design project where I need to feed a reactor with 4000 kg/h of propylene at 20 bar abs and 250 degC and I have been searching different alternatives regarding storage and vaporization of propylene.

Looking around books, the web and this forum (mainly posts regarding LPG storage and vaporizers) I understand that a suitable layout would be to have a vaporizer and a superheater downstream. The vaporizer would operate at 21 bar (1 extra bar of pressure as driving force) and 51 degC temperature (saturation value @21 bar) and the superheater will get the vapor up to 250 degC to feed the reactor. According to NIST data the heat of vaporization @21 bar is 270 kJ/kg so for my propylene flow the heat duty of the vaporizer would be 305 kW. The heating source is low pressure steam.

I have read that propylene vapor is send to the consumption point (the reactor in this case) from a tank that has propylene with both liquid and vapor in equilibrium (@21 bar in this case) and the liquid is sent to a vaporizer that recycles the vapor formed back to the tank to be consumed. The pressure in the tank is going to be controlled with the heating source input and the liquid vaporized will be replenished from a storage tank by a level controller.

Regarding storage, my intention is to have bullet tanks at ambient temperature (25 degC) and 22 bar (another extra bar). In this way, I will have the propylene readily available with enough pressure to be sent to the working tank and vaporizer. I read that in some cases storage in some cases is done at low pressure and the tank has a refrigeration system but, since I need the propylene at high pressure and temperature for my reactor I consider the first option is better.

The questions I have so far are:

- Do you think the vaporizer/superheater system I described is a good option?
- What are the usual vaporizer equipment in the field? Shell & tube heat exchangers? Which TEMA type?
- The storage selection is acceptable or another option is more suitable?
- The pressure differences will be enough driving force for the propylene to flow or pumps/compressors will be needed in some point?
- Which heat exchanger is most suitable for superheating applications?

I am on the basic design stage so some basic guidelines will be sufficient. I am attaching a drawing sketch adapted from a book with some comments to clarify.

Looking forward to read your insights. I appreciate your help in advance.


Regards,
damik

Attached Files

•  propylene vaporizer sketch.png   34.1KB   11 downloads

[Pilesar]

Keep your bullet tank 'just a tank' without internal coils. Bullet tanks for propane are very common. Propylene tanks are similar, but require slightly higher pressure than the usual rating for propane tanks.

 

There is a tendency to want to combine functions to save equipment cost, but keeping simple equipment with one function is often cheaper in the long run. It is also easier to design single function equipment!

 

You can gravity feed an external heat exchanger to return vapor to keep your bullet tank pressure up if you need it. With an external heat exchanger, you have better access for maintenance and flexibility of operation.

 

You cannot count on bullet tank pressure being high enough to overcome pressure drop in lines, control valves, etc, to reliably feed your downstream reactor unit. The reactor feed rate should be constant. Your bullet tank pressure will vary with the ambient temperature and with in-flow and out-flow. Add feed pumps for the liquid propylene from the bullet tank. These pumps and cost little but give much more reliable feed rate to the reactor.

 

Pump the liquid propylene from bullet tank to the vaporizer and superheater. Measure and control the liquid flow rate to keep a constant mass rate to the reactor. You will want the pump discharge pressure high enough to avoid two-phase flow in the line. The transfer line for liquid will be smaller than if you are transferring vapor. If you vaporize and superheat close to the bullet tank, you will lose temperature in the long transfer line and may condense some of the propylene. The vaporizer and superheater should be close to the reactor and you don't have to try and keep a long vapor transfer line hot.

 

Spend more time evaluating the process flow sheet before designing the heat exchangers. Besides researching propylene vaporization, also research propane vaporization and LPG vaporization as the equipment is similar.

[damik]

Thanks for your comments Pilesar. I will consider your inputs and keep investigating.

 

 

Regards

[Art Montemayor]

I have done this type of vaporizer application before, so I’ll add some comments to Pilesar’s excellent reply.  Everything that he has stated is, in my opinion, the correct and simple way to analyze and design a vaporizer for this type of application.  The only item I would comment on is that my experience with bullet LPG tanks has shown them to be designed with a nominal 350 psig MAWP (approx. 25 barA).  Nevertheless, his points are well taken.  I would not employ the Propylene storage tank as a reservoir for saturated pressure at 22 or 25 barA.  This means that you would have to maintain the storage tank at 53-60 C.  Don’t forget: you are storing saturated liquid.  Get used to that fact of life and learn to interpret the related thermodynamic properties as shown to you on the attached workbook.  Keep everything as SIMPLE as possible - but always SAFE.  Minimize flanged or gasketed connections wherever possible.  NEVER, never use screwed fittings on a highly flammable fluid such as propylene.

 

Check my calculations and tell me if you see anything wrong with my results.  Don’t be shy or afraid of challenging an engineering calculation or data.

 

Since it is not practical to maintain your storage at an elevated temperature (how you could do it is up to your imagination!), you should scrap that idea and follow the simple, direct, and common sense of getting a higher pressure propylene to vaporize and subsequently superheated.  That makes a lot more sense and is more controllable.  This also makes more sense from an operability point of view because it allows you to have backup equipment such as pump and vaporizer with the ability to isolate and replace/maintain.

 

It is important that you identify the approximate size of the required transfer pump between the storage and the vaporizer.  This should indicate to you what type and size of pump you should apply and its physical size.  This has to be continuously pumping as the vaporizer and superheater are working.  It is also important that you know the vapor rate going to the superheater - for obvious reasons: you need to establish the size of vapor space you need in the vaporizer that serves as the reservoir feeding your superheater and reactor downstream as well as the size of the superheater.  You don’t want your vaporizer’s vapor space pressure fluctuating up and down because of a lack of sufficient vapor volume with the liquid level varying in the vaporizer.  You control the liquid level in the vaporizer such that it covers the heating element and the degree of level variance can affect your level control valve’s delivery makeup rate.  Design your vaporizer much in the style of a BKU TEMA kettle reboiler EXCEPT: try to make the tube bundle as a fixed, welded unit - much like injecting the low pressure steam into tubes that have their inserted ends welded and are themselves welded to a tubesheet.  This design allows for a hermetic heating system it is highly unlikely to have any propylene leaks.  Since you have very clean fluids on both sides of the vaporizer bundle it makes no difference whether the bundle is fixed or not as far as maintenance is concerned.

 

The superheater is another device close downstream of the vaporizer and it can be a TEMA U-tube design - with one process exception to your statements: You CANNOT use “low pressure” steam as the superheater heating fluid.  Once again, study your thermodynamic properties and always refer to them and check out whether the desired heating can actually take place with a reasonable degree of operability.  Again, refer to the steam thermo data in the workbook attached.  It is good that you are using NIST as a source for therm data - but you have to pay attention to the practical value of the data given and what it tells you.  Low pressure steam simply can't furnish the superheated temperature you are seeking.

 

I started to draw a simplified (and very useful) flow diagram, but I have run out of time and actually this should be YOUR responsibility and job.  Pilesar and I have already given you more than sufficient input for you to nail the application and obtain a very high grade in this exercise.  Good luck.

 

 Propene & Satd Steam Properties.xlsx   113.03KB   61 downloads

 

[damik]

Dear Pilesar & Art,

 

First of all, thank you very much for taking the time to help me giving such answers with lots of helpful information. Sorry for the late response, I have been reading carefully your comments and I have some questions I would like to ask:

 

• You are both recommending a pump to send the liquid propylene to the vaporizer. Since the liquid is saturated, would not the pump be susceptible to have cavitation issues? Is there necessary to have a special pump for this application (like low NPSH pumps)?
• Regarding the previous question, would be beneficial somehow to have the propylene subcooled and not saturated?

Some more things I would like to clarify:

 

• My intention is to have the storage tank and the vaporizer separately. I am seeking for a layout similar to one I saw Art recommending in another post in which the tank (named working day tank) has a recirculation line with the vaporizer in it and the vapor formed is recycled to the tank in order to be sent to the superheater/reactor.
• For the superheater I am not thinking of using steam since I would need to use high pressure steam, at least 40 barA. I considered this option, also using a fired heater but right now I am considering using some thermal fluid. I would like to know what are your thought about this option.

Once again, thank you very much for your valuable help.

 

Hope to hear from you again.

 

 

Best regards

[Art Montemayor]

Damik:

 

Since the liquid propylene is saturated, the pump is susceptible to having cavitation issues if you fail to take them into consideration.  Show your instructor/grader that you know what you are doing and that you are designing a safe, operable operation by showing that you know the size of the pump, the type of pump required complete with flow control, and reveal your detailed NPSH calculations.  I always tried to have my LPG tanks’ bottom nozzle at least 6 feet above grade.  Use a pump suction line at the very least 1 pipe size larger than the pump’s suction nozzle and use as low of a pump rpm as you can get away with.  If designed and operated properly, you should not have cavitation or flashing problems.  Liquid propylene is quite like LPG - except in density I believe.  Study the attached related literature on the type of pump(s) commonly used in this type of application.

 

I would recommend a thermal fluid for the superheater - for thermal and safety reasons.

You require a pump discharge line return to storage tank for capacity control purposes.  Again, refer to the attached literature.

 

It is far more economical and simpler to use a pump instead of a compressor to get the higher pressure, superheated vapor you want.  Quit fooling around with having to raise the entire storage tank’s temperature to obtain the saturated liquid conditions you need to avoid a pump (or compressor).  It doesn’t make for practical horse sense or economical operability.  You simply cannot return vapor back to the storage tank without having to employ a compressor to raise its pressure.  What are you thinking?  Maybe I’m wrong because you still haven’t even generated a simple detailed flow diagram, but I think I understand your process.  The vaporizer application drawing I did was on liquid CO₂, I think.  And that’s a totally different application where the ultimate downstream vapor consumption is well below the storage tank’s saturated pressure.  Am I correct?

 

The main scope of work in your design is a change of phase with a required product pressure well above the saturated liquid.  It is much simpler to use a pump to get to the higher pressure as liquid and then vaporize it and superheat it.  If you are interested, do the hp requirement to pump the liquid as well as that to compress the equivalent vapor - and then compare both.  

 

 Blackmer LPG Pumps.pdf   4.3MB   33 downloads

 Conquering the Cavitation Conundrum.pdf   3.24MB   26 downloads

 SIHI LPG EN A4.pdf   1.39MB   29 downloads

[damik]

Dear Art,

 

The information you shared about the pumps (and the pumping system design) will be very useful and I really appreciate it.

 

The post I mentioned previously was one about propylene vaporization also (https://www.cheresou...nd-air-heating/) where you attached a sketch of your recommended layout (Ahmed__s_Propylene_Storage___Vaporization.xls) in which there is the propylene storage tanks farm that feed the "working day tank" with a vaporizer in its recirculation line.

 

I was thinking of using a similar system but, now I read that post in more detail, the storage pressure (equilibrium pressure at ambient temperature) was higher than the pressure needed in the reactor. In my case, I need higher pressure than the storage temperature so that layout is not valid and the pump is necessary to obtain the pressure desired for further vaporization and superheating of the propylene. Am I correct?

 

P.D. I put the links of the previous post and not attached the file in order to avoid uploading the same information twice.

 

 

Best regards,

damik

[Art Montemayor]

Damik:

 

I hate being a strict interpreter of terms employed, but I consider it important in this topic because it involves phase equilibria - a subject vital for understanding the basis of your storage and vaporization needs.  Therefore, I have to be critical of your description of the type of unit operation you now will undertake.  You state, “the pump is necessary to obtain the pressure desired for further vaporization and superheating of the propylene”.  Your statement is basically in error.  The propylene pump is necessary (in the flow diagram you still have not generated) because:

• The saturated conditions (temperature and pressure) existing at the proposed ambient temperature for the liquid propylene are below the target conditions designated for the downstream reactor.
• Consequently, an intermediate vessel/vaporizer is employed to receive liquid propylene from the pump at a higher pressure and temperature by applying heat of vaporization in that same vessel.  This intermediate vaporizer vessel maintains its contents in the saturated state as the vapor propylene is drawn out by needs downstream.  As the liquid is depleted by vaporization in this vessel, the transfer pump, acting on level control, maintains the vaporizer liquid level.
• The higher vapor pressure facilitated in the vaporizer allows for positive vapor flow out of the vaporizer, through the required superheater, and into the target reactor vessel.  The heating fluid in the vaporizer (low pressure steam) is pressure controlled to maintain the vaporizer vapor space at the continuous, desired saturated pressure.
• The main propylene storage tank is maintained at normal, ambient temperature and its corresponding saturated pressure.  This operating condition allows for normal makeup propylene to be added routinely to the storage tank under operating conditions or when the reactor process is not working.

Note that I have stressed the importance of understanding and employing the phase equilibria existing in both the storage and vaporizer vessels.  This is key to allowing you to pump a liquid and subsequently vaporize it at the higher pressure you require for superheating and injecting it into your reactor.  You do not have - as you state - “further vaporization”.

 

I hope these comments help you proceed forward in producing an engineering design and report you can be proud of.

[damik]

Dear Art,

All your comments in the last post were absolutely clear and I was not in my last answer. English is not my native language and I may have commited an error using some words in the wrong way.

 

I really appreciate all your help and all the knowledge you have shared. This will help me to move forward with the design.

 

 

Best regards!

[Art Montemayor]

Damik:

 

No hay de que.  Buen provecho.

 

I'm sure that Pilesar joins me in wishing you well and that this lesson learned helps you obtain a grade deserving of the work input you put into your project design.  All I did was add some details to the basic, good design advice he supplied.