17 replies to this topic

[Geetanjali]

Hi everyone,

 

Please see the attached image. I recover gases (H2, CH4, CO) from the top of col DIST-1 at -118 degC and 7 bar. I have to bring this to room temp and ambient pressure. I also want to save energy in doing so. 

 

In order to do so, I first installed a valve that brought the pressure down to 1 bar but also further reduced the temp of stream 404 to -125 degC. I now use a heater to simply heat this stream to 25 degC.

 

Does this make sense? Look at the heat duty required by heater (HX-400-2) = 6013 cal/sec. 

 

Instead should I use other approach? For example, using this low temp stream to advantage rather than simply wasting it?

 

 

Note: I am separating a stream of gases to recover C2,C3,C4 olefins using a de-ethanizer, de-propanizer and de-butanizer type columns. The DIST-1 col is the first one in the series that will remove other gases (H2, CO, CH4) from these olefins.

 

Thank you,

Geetanjali

Attached Files

•  Problem.JPG   77.69KB   2 downloads

[breizh]

Hi,

Will it change much if you install the reducing valve after the HX instead?

my 2 cents

Breizh 

[Pilesar]

In commercial olefins plants, the 'Dist-1' demethanizer overheads are used to cool the column feed. There are many examples of olefins recovery by distillation. A little bit of literature search should yield quite a lot of information.

[Geetanjali]

@Breizh: Thank you Breizh, Indeed the valve wasn't doing much when that heater simply could do both job (Temp rise and Pressure reduction). Thank you!

 

@Second responder: Thank you! Indeed that's a great tip. I am referring to several literature reports including books (by Robert Myers, Kirk-Othmer Encyclopedia of Chem Technol and others). I do understand this portion requires a very good heating and cooling integration especially because I am using refrigerants to cool some streams (in olefins recovery columns). I am planning to learn pinch analysis (HIN) next and I hope that'll give me better perspective and understanding of using  utilities wisely.

 

Thank you both of you again, this is really helpful!

 

Best,

Geetanjali 

Edited by Geetanjali, 28 July 2021 - 10:33 AM.

[PingPong]

 I do understand this portion requires a very good heating and cooling integration especially because I am using refrigerants to cool some streams (in olefins recovery columns).

 

The heating and cooling integration of an olefins plant is much more complicated than you can imagine, especially if you want to operate the Demethanizer column at only 7 bar. It is possible but that complicates the design severely as you will then need a methane refrigerant compressor or or tertiary refrig system or a turboexpander.

Why did you chose this low Demethanizer pressure?

I have worked on several olefin plant designs in the past and because you have no experience with this I suggest you chose a considerably higher Demethanizer pressure.

 

A few more remarks regarding your flow scheme:

 

Why is there a compressor MCOMPR-C in the liquid stream from the Demethanizer bottoms, apparently resulting in a vapor and a liquid stream? That can't be right.

 

The Demethanizer bottoms liquid is to be pumped via the heat/cold integration (a.k.a. Chilling Train) to the Deethanizer column. The distillate from the Deethanizer is then sent to the Ethylene Fractionator to separate the ethylene from the ethane, which is recycled as feedstock to the cracking furnaces.

[Geetanjali]

Dear Sir/Madam,

 

Thank you so much for looking over the snapshot of my Aspen Model. I have incorporated your suggestion and raised the pressure of Demethanizer column. Please see the snapshot of my model attached.

 

Previously, I used 7 bar as I was referring to a thesis (https://digitalcommo...u.edu/etdr/797/) that used this pressure in their de-methanizer column. However, they had a detailed refrigeration cycle model which I am not doing in my process. Thus, it makes sense to not go low in pressure, hence I am using a greater pressure. Thank you! By doing that, I can go without a multicompressor that I was previously using (I had removed it after posting my first post because it was unnecessary). 

 

Indeed this is my first experience of working with olefin plant and I welcome all the ideas. I am very thankful to you for giving your time and valuable suggestions. 

Attached Files

•  Model.JPG   52.63KB   0 downloads

Edited by Geetanjali, 29 July 2021 - 02:23 PM.

[PingPong]

That thesis is about PYROLYSIS OF WASTE POLYOLEFIN PLASTICS according to its title, but I can't see the full thesis text.

 

I assumed that you were designing a complete olefins plant but apparently not, it is something simpler, right?

 

 

What exactly is the description of your task?
 

[Geetanjali]

Hello,

 

I am designing a plant that makes olefin gases and aromatics rich naphtha from the pyrolysis of plastic waste. I am trying to recover olefins (mainly C2, C3 and C4) by the condensation and extraction of the gaseous stream that's why needed to design olefin plant portion. 

[PingPong]

So your starting point is the effluent from the pyrolysis section, right?

 

After cooling that will have a temperature of 40 oC or there about, I would expect. What is the pressure at that point?

 

You show a distillation column to separate ethylene and CO2 but that is not the way to do it. If the pyrolysis effluent gas contains CO₂ then that has to be removed by absorption in a solvent before the gas enters the chilling train.

 

The choice of pressure in demethanizer and other columns depends on the choice of cooling medium for condenser and heating medium for reboiler.

 

Condenser cooling is in some columns some sort of refrigerant so you need to think about that first.

 

In an olefins plant there are multiple demethanizer feed streams coming out of the chilling train, at different temperatures and with different compositions, and fed to different locations in the demethanizer.

 

The demethanizer condenser is the coldest point in the plant. You have to think about how to achieve that, and in combination with that decide on the demethanizer pressure. It's a trade-off: the higher the demethanizer pressure the less power required for refrigeration and the simpler the design becomes, but the more power is required for the compressor to get the effluent gas to the pressure in chilling train and demethanizer.

 

Pinch analysis is not really of much help in these conceptual design decisions, although academics at universities who never designed anything that actually got built and performed satisfactory will probably disagree with my opinion (which is based on 30 years of actual design experience).

[Geetanjali]

Hello,

 

Thank you for your valuable insights and great suggestions. Please see the snapshot attached my entire model. I am using mixed waste plastics stream and converting into pyrolysis naphtha (rich in aromatics) and gases (rich in olefins) in a catalytic fast pyrolysis reactor. Area 300 - Distillation separates these two product streams. Area 400 - Olefins separation is where I am trying to recover C2,C3 and C4 olefins from the gaseous stream. A500 - Aromatics separation - is where I recover BTX aromatics using UoP's sulfolane process.

 

1. The pressure after cooling the pyrolysis stream and separating naphtha (liquid) and gases (olefin rich) is 21 degC.

 

2. Thanks for suggesting about CO2 separation using absorption columns. Indeed, what I was doing may not be relevant in real world (but work in paper). I would try to have an absorption column before it goes to chilling train. Do you have any inputs on that? Should I use the CO2 absorption column before or just after demethanizer? How can I share my aspen model with you? (I can attach it here, but I dont want to do it here)

 

3. I am using utilities (refrigerants) for the columns in Aspen Plus. mainly propane, ethylene.. Do you have suggestions on refrigerants too?

 

4. I really want to do a simple run without much optimization especially for recovering olefins from the gases. I understand it is non-trivial. But, I have no experience of how it is done in the real world. I really appreciate your time and inputs. I am learning a lot.

 

 

Sorry for asking too many questions

 

Best,

Geetanjali

 

 

 

 

Attached Files

•  Model.JPG   82.84KB   0 downloads

[PingPong]

1) so what we are talking about is A400 which gets a gas feed at 21 oC and 1 bar(abs) right? I wonder how you achieve that 21 oC.

 

2) CO2 removal is upstream the chilling train. Demethanizer is downstream the chilling train.

I attached a simple flow scheme of a typical olefin plant to give you some idea.

The chilling train consists of E-511/512/513/514/515.

 

For the moment it is not important how the CO2 is removed. Just put a block in the simulation that takes it out and give it some pressure drop. Later, when the important design work is done, you can design the CO2 removal.

 

I don't use Aspen so don't share your model.

Moreover the model will change every day as you are still learning, so outsiders would not be able to keep track of all that.

 

3) ethylene and propylene, or propane, are often used in olefin plants, especially the older ones. Newer plants often have binary or even tertiary refrig systems and sometimes also turbo-expander(s). Choice is up to you.

 

I notice that the flowrates (kg/h) are very small so in the real world that would not justify a lot of extra equipment to save energy as there is little energy to be saved in your small plant.

 

 Typical olefin plant flow scheme.jpg   415.7KB   0 downloads

Edited by PingPong, 30 July 2021 - 02:36 PM.

[Geetanjali]

Thank you again! I referred this figure and googled to get the entire paper. As suggested, I am using a simple separator block upstream and taking out the CO2 from there. Next I am inputting CO2 minus stream first to a demethanizer column at 37 bar (3.7 MPa) that completely separates H2, CO and CH4. The rest goes through the De-ethanizer column and C2-splitter (that now gives me ethylene and ethane separation). Bottom stream 407 goes to de-propanizer. I feel much confident now. This was the first pass and it is giving me good separations.

 

I will further play with the column specifications to achieve >97% purity of these olefins. But this was really really helpful. I would have never known that we should not put CO2 down through the chilling train - a great tip. Also, about the utilities requirement using a demethanizer at low pressure vs that higher pressure. Thank you again. 

 

To answer your other question.

 

1. I was getting the pyrolysis reactor effluent at 635 deg C. I used chain of coolers and Flash 2 vessel to separate C1-C4 gases from liquids (stream 305). See attached image

 

I have also attached snapshot of where I remove CO2 using a simple separator and the rest of the stream then goes through chilling train especially first through demethanizer, followed by de-ethanizer, de-propanizer and finally de-butanizer.

 

 

I am very thankful for your time and great suggestions. This is really helpful and I learnt a lot especially why we don't use certain unit operations in a particular sequence. Indeed what might work well in paper, might not in real world. Grateful to receive inputs from you who has far greater experience of some real industrial experience.

 

Best regards,

Geetanjali

Attached Files

•  olefins sep.JPG   73.37KB   0 downloads
•  co2 removal.JPG   51.62KB   0 downloads
•  A300.JPG   46.09KB   0 downloads

[PingPong]

Your simulation of the compressor stages is not correct.
The first compressor COMPFEED increases the pressure from 1 to 5 bar. That is not possible as that would result in a far too high discharge temperature resulting in excessive compressor fouling.
The compression from 5 to 37 bar is done in MCOMPR1 resulting in a vapor stream and a liquid stream at -90 ^oC. That does not make any sense either. That is not a proper way to simulate multiple compressor stages and chilling train.
In an olefin plant it takes 5 compression stages to reach 37 bar. After each stage there is a cooler and a separator (KO drum) to separate condensed liquid which is sent to the previous KO drum. The cooled and dried vapor from the fifth compressor stage flows to the chilling train.

 

I browsed through the Alvarez 2019 thesis that you mentioned. There is no demethanizer in their system. The methane plus helium is separated in condenser E-240D at -136 ^oC & 7 bar.
I surprises me that that simple single stage flash would be enough to get sufficient separation between methane and ethylene. There is no detailed material balance with a component composition of all streams so it is for outsiders impossible to check anything. Apparently the persons that reviewed and approved that thesis did not care about that sort of details.

 

 

What is the detailed composition of the feed gas to your unit A400, I mean stream C1-C4,C5(IN) ?

Edited by PingPong, 31 July 2021 - 03:36 PM.

[Geetanjali]

Thank you again for your valuable inputs. I have now removed the single multi-stage compressor I was using earlier and now have 5 compressors followed by coolers and 5 flash2 drums (mimicking KO pot) to take out the condensed liquid (please see attached image). Please let me know if this sort of compression and condensation train looks correct?

 

Thanks for checking the Alvarez thesis. Since it is an academic exercise, I assume the authors did not give details to its applicability in the real life and hence used single distillation tower to bring the separation of CH4 with ethylene.

 

I am also attaching a snapshot of the stream composition of C1-C4,C5 (IN) for your reference.

 

 

Edit: Please note - stream C1-C4,C5 also contains some naphtha component that gets separated in Col SEPC5 and olefins separation train only receive Stream 400 for which I have attached the stream composition.

 

 

Thank you for your time and valuable insights.

 

Best,

Geetanjali

 

Attached Files

•  STREAM GOING TO OLEFIN SEPARATION.JPG   62.36KB   0 downloads
•  stream 400.JPG   94.4KB   0 downloads
•  37 bar.JPG   61KB   0 downloads
•  c1-c4 and c5.JPG   86.72KB   0 downloads

Edited by Geetanjali, 02 August 2021 - 06:40 PM.

[PingPong]

I am puzzled by what you are doing.

Before I already questioned the 21 oC. Usually that is not achievable with cooling water only. It depends of course where the plant is located, and what kind of cooling water is used: closed CW, river water, sea water. You should think about what the supply temperature of your cooling water is. Return temperature is then typically 10 degrees higher and so is usually the outlet temperature of the watercooled exchangers. For example: if your CW supply temperature is 25 oC then you can cool the process gas to about 35 oC with that.

 

You seem to use refrigerant to cool the compressor interstages

to 10 oC in COOL1

to -10 oC in COOL2

to -30 oC in COOL3

to -60 oC in COOL4

The drawings are difficult to read so maybe I misread some numbers.

 

Nobody would do it like that in the real world. There the interstages are cooled by cooling water only, and the liquid condensate from FLASH5 is sent to FLASH4, and liquid from FLASH4 to FLASH3, et cetera.

 

What is SEPC5 ? Is that a reboiled stripper with stages but without condenser? Or is that exchanger a cooler?

Is that pressure of 5 bar determined by optimization, or is it an arbitrary choice?

 

You will need 2 compressor stages (not one) to increase gas pressure from 1 to about 5 bar, followed by 3 (not four) stages to further increase pressure to about 37 bar.

 

Note also that there is to be a pressure difference between discharge of a compressor stage and suction of the next due to pressure drop of exchangers, separators, CO2 removal and piping.

Edited by PingPong, 05 August 2021 - 08:51 AM.

[Geetanjali]

Thank you again for all your valuable inputs and apologies for responding late here. However, I was studying more and more about it and now have re-drawn my entire olefins separation flowsheet in Aspen Plus. I have attached the flowsheet in 2 parts since it was big.

 

Please note:

1. "Olefins" stream enters the flowsheet and is compressed and cooled to about 38 bar and 50 C.

2. It then is cooled by the several chillers(using refrigerants) and further cooled by using a heat exchanger at several locations in the flowsheet.

3. After this cooling (by chiller), cooling by heat ex, the stream is flashed to separate liquids from the vapor streams using flash drums (F1, F2 and F3).

4. These liquid streams then enters at multiple stages in the De-methanizer column (1DEMETH).

4. The top stream from the 1DEMETH (1DIST) is mainly methane and Hydrogen, is first expanded using a Expansion Device (EXP) to lower its pressure down to ~4.5 bar and then used subsequently in the flowsheet and finally recovered via stream CB2-PROD

5. The bottom stream from Demethanizer (1BOT), then enters Deethanizer column where C2's are recovered from the top (2DIST) and bottom stream then goes to a De-propanizer column.

6. The top from Deethanizer column containing ethylene and ethane is separated in a C2 Splitter (3ETHFRAC)).

7. Propanizer column, 4DEPROP separates C3s in the top stream (3DIST) which goes to a C3 splitter (5PROPFRAC) to separate propane from propylene.

8. The bottom stream from 4DEPROP then enters a debutanizer column, 6DEBUT and separates C4s in the top and C5+ in the bottom.

 

Thank you and regards,

Geetanjali

 

Attached Files

•  PArt 2.JPG   58.02KB   0 downloads
•  PArt 1.JPG   63.99KB   0 downloads

[PingPong]

In actual practice streams in the chilling train are cooled and reheated counter-currently in Plate-Fin Heat Exchangers also known as Brazed Aluminum Heat Exchangers (BAHX). Google that.

 

Multiple BAHX with insulation around them are combined in a so called Cold Box, which sometimes also includes vessels.

 

Usually in a process simulator there is a type of heat exchanger that can be used to simulate such exchanger. I don't know how it is called in Aspen Plus. In PRO/II it is called an LNG exchanger, maybe also in aspen. Check your user manual.

 

In such LNG exchanger multiple streams, including refrigerants, can be handled counter-currently. Usually a Minimum Temperature Approach (MITA) of 3 ^oC is specified by the user. MITA should also be mentioned in your aspen user manual, I would expect.

 

Also realize that heat leak into cold streams is a factor to take into account as it increases refrig requirement. I would expect that also in aspen one can specify for each BAHX the heat leak as percentage of duty. Check your aspen user manual.

Edited by PingPong, 23 August 2021 - 04:41 AM.

[Geetanjali]

Thank you for your valuable insights. These will be very helpful during the economics assessment of this plant. I have checked, we use shell and tube heat exchanger in aspen plus with several configurations possible for the co-current or countercurrent flow and likes. Yes, I have used a minimum temperature approach of 3 deg C while simulating the Heat exchangers.

 

Also, thanks for pointing towards heat leak. This is a great pointer and will include it in my economics calculation as a percentage. Will check the aspen manual for more details.

 

Thank you again. I truly appreciate your help. I learnt a lot from this exercise and feel really good about my aspen model and I am presenting it tomorrow to my mentors. 

 

 

Thanks and best regards,

Geetanjali