Jump to content

Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Gasoline Blending- posted in Ankur's blog

Effects Of Pressure On Distillation Energy Requirements

This topic has been archived. This means that you cannot reply to this topic.
15 replies to this topic
Share this topic:
| More

#1 Bennett Willis

Bennett Willis

    Brand New Member

  • Members
  • 6 posts

Posted 01 March 2008 - 04:08 PM

I'm an industry retiree (33 years as a chemist doing process research and plant support) and am training operators at a community college. In one of my classes, we are having a discussion about whether running a standard distillation (separating a feed into overhead and bottom streams of different compositions) would take a lot less steam (25-50% less) if you ran it under a vacuum--and thus the energy savings would motivate you to run something under a vacuum. I don't think that it makes enough difference to cause people to run under vacuum and my class says that this is not what they have been told. I promised them that I'd try to find some information on the topic and this is part of that effort.

What I saw was that you ran the pressure on distillations to meet the needs of the overheads (condensing temperature/energy recovery) and the bottoms (product decomposition, energy recovery from a process stream or steam pressure available). Vacuum separations seemed to require larger columns due to the higher vapor velocity caused by the much larger volumes of vapor due to the low pressures and you had to have a vacuum system. The only vacuum columns I had experience with were run under vacuum due to problems with bottoms temperatures (generally product decomposition or color development). I think that I remember a few cases where the column pressure would be changed so the column could recover heat from a process stream.

From my chemist's point of view, it should take about the same amount of energy (pounds of steam or BTU's) to do the same job (roughly) regardless of the pressure you are doing it.

My position is "strongly held but poorly documented." Experience has taught me to be ready to correct my model and I am prepared to do so in this case--should it be appropriate. Could some good engineer offer some intuition to a chemist?

I would need some discussion on this as I have been telling my classes that it takes roughly the same amount of energy to do the same job "regardless of the path." This puts me in conflict with what they have been told in other classes. I don't feel too bad about the conflict because none of us know about how the energy required changes with pressure. I suppose that I feel that I am closer to right than the other instructors because vacuum columns were really rare where I worked except for issues related to bottom temperatures. This should not have been the case if the amount of energy drops significantly when you reduce the pressure.

Bennett Willis
Brazosport College

#2 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 5,490 posts

Posted 01 March 2008 - 05:30 PM


Welcome to our Forums. I have taken the liberty of moving your thread to the Industrial Professionals Forum because I don't believe this subject matter is one that can reasonably be expected to be handled by chemical engineering students. We experienced chemical engineers have our problems with distillation as it is. But I think exposing your thread to the experienced and professional guys, you should get more experienced exposure and some direct and positive responses.

I hope this is OK with you and that you obtain the results you are looking for.

#3 Zauberberg


    Gold Member

  • ChE Plus Subscriber
  • 2,692 posts

Posted 02 March 2008 - 09:33 AM

Hello Bennett,

Posted in this manner, your question is very similar to Joe's recent thread regarding steam consumption of distillation tower at elevated operating pressures - it as almost 100% academic subject. Without knowing any details about particular service/application, I find it to be impossible to answer all your queries, simply because we don't know what are we talking about: is the simple phase transformation of boiling component (or mixture) in question here, or you are refering to industrial fractionators and their operating envelopes. Furthermore, what parameters we consider to be fixed: feed enthalpy, reboiler heat input capacity, condenser heat removal capacity, pumparound loops configuration, tower height/diameter, product specs, environmental/safety issues etc etc. As you can see, there is a lot of things here which need to be defined and resolved, before we can move forward and bring the conclusions.

I spent 5 years of my professional life in operating, troubleshooting, modeling, revamping and inspecting industrial fractionation equipment. I became familiar with many beautiful subjects related to distillation columns, and had many occasions to resolve personal dilemmas which I was carrying with myself all the way back from university days. In my humble opinion, this is the perfect place where you can find a lot of people which can help you to solve your mysteries - if any. Sjoerd has already gave you few brilliant answers in Student forum. Also, there is a number of free resources available on the internet which can provide you a guidance for distillation equipment design and operational concerns. But, I think all details of your particular application should be known in order to start quality discussion without shifting to theoretical levels.

Welcome to ChE forum and hope you'll enjoy spending some time here.
Best regards,



    Gold Member

  • Members
  • 57 posts

Posted 03 March 2008 - 09:18 AM


As what I learned from years of production, lowering pressure(without causing flooding) is a major measure to save energy consumption in a specific distillation column. However, that how much may be saved depends on many factors.

I would recommend you to read a linked article for some reference( http://www.chemicalp...s/2007/090.html ), and also in my post, Zauberberg and Ali66 has same conclusion from differerent explanation.


#5 Bennett Willis

Bennett Willis

    Brand New Member

  • Members
  • 6 posts

Posted 03 March 2008 - 01:55 PM

Thank you Art. I'm still working on understanding the logic of the site.

What I am finding when I look around the net is that there is almost no one who says that you routinely expect significant energy differences (over a few percent) by dropping/changing the pressure. I've asked our library to see if they can borrow the book [Shinskey, F.G., “Distillation control for productivity and energy conservation,” McGraw-Hill, New York (1977)] that was referenced in Joe's note. Shinskey reports 25% energy reduction with pressure change under some circumstances and I am curious as to whether he has just gone from really bad operations to really good operations or what.

Our particular column is a training unit (15 inch diameter, 8 sieve trays) and runs under a pressure that generally ranges from 250 to 90 mmHg. We are separating water and ethylene glycol and make relatively pure water overhead and bottoms that are enriched in glycol. We are limited to about 18 psig maximum steam pressure and 90 mmHg minimum operating pressure. I'm trying to come up with a general statement for how you decide where to operate your distillation column. The conclusion that I am coming to is that you set it up so that the overheads condense and the bottoms vaporize and don't "burn" and consider heat recovery from both tower overheads and from other vapor streams that might be available to drive the tower. I'm trying to see if I really need to say more.

Bennett Willis

#6 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 5,490 posts

Posted 03 March 2008 - 03:40 PM


I think you may get some input from Greg Shinskey in his book. He usually is very good about pointing out effective and optimum methods in controlling processes – especially distillation. However, I believe you will come to the usual situation where you have to define what is the scope of work of the subject distillation column and one of the immediate conclusions is that you can’t generalize and be correct all the time.

Have you checked out Andy Sloley’s webpage?: http://www.distillat...com/distill.htm

He runs his consulting services out of College Station, I believe, and he undoubtedly has come to a definite conclusion with respect to this specific query. Andy is a world-recognized expert in distillation and I don’t believe there is a situation that would be strange to him or that he hasn’t already resolved in the past. At the bottom of his Distillation webpage:


You will find the download reference, “Texas Energy Conservation Program Distillation Column Operations” which you and I have paid for with our taxes and which is a long and interesting report on the economics and energy savings to be found in a distillation operation. You may already have this information for your class, but I mention it just in case. I don't believe Andy has visited or written in our Forums yet. I wish we could be honored with at least one of his visits.

I have not had the opportunity to vary my operating pressure in a distillation column design or operation. Every design and operation I’ve done has had basic operating pressure requirements due to the process linked to - or part of – the distillation section. Therefore, I usually have always looked for the lowest possible positive operating gauge pressure (& temperature) I could get away with. As long as I had good volatility difference in my components, I have been OK. I’ve also done vacuum – and there hasn’t been any option there except to keep the pressure drop as low as possible across the column.

The obvious trade offs when I have considered lower pressure – and even partial vacuum operation is that the column diameter immediately starts to increase – due, of course, to the higher specific volume of the ascending vapors plus the constraints of keeping the pressure drop low. When you are dealing with a stainless construction, it can really start to affect the capital costs very rapidly.

I hope some of our experienced professionals continue to contribute to this thread.

#7 Bennett Willis

Bennett Willis

    Brand New Member

  • Members
  • 6 posts

Posted 04 March 2008 - 10:25 AM

I did find a Vulcan Chemicals article showing a 5% savings in steam by lowering the pressure in a couple of columns (from 35 psig to 28 psig and from 15 to 11 psig). They had some "room at the top" and could just reduce the operating pressure. No equipment changes were required. These are the "free money" ones.

I am convinced that running a lower pressure is generally a move in the right direction for an incremental improvement in steam consumption. As Joe and Zauberberg both pointed out, you should select the lowest design pressure you can consistent with all the other needs of the system. I liked Zauberberg's statement and have quoted it here for emphasis. "Operate the tower at lowest possible pressure at which you can achieve required condensing duty. That's the basic rule of distillation and it will not change as long as this world exists. Lower the column operating pressure and you will need less steam to achive the same process (quality) targets." Thus we should all walk out into our plants and see if there is any tower that can be run at a lower pressure without having to modify the equipment.

I thought that a comment I found somewhere about not "over purifying" was a good point. I recall one case where the amount of a recycled impurity being made in the reactor turned out to be very close to the specification amount of the impurity if all the impurity was in the product, but we did not know exactly what the make rate was. The plant routinely produced material that was about 50 ppm below spec--a typical behavior. Over a few weeks the impurity would build up in the system to several % and keeping it down to the normal level in the product required a lot of extra reflux. This loaded up the column and actually was the production limiting issue as well as representing a large amount of energy. This also gave the general impression that there was a lot of the impurity being made in the reactor. When the amount of the impurity in the product was increased to the actual specification, the impurity disappeared from the plant recycle stream.

Bennett Willis

#8 Zauberberg


    Gold Member

  • ChE Plus Subscriber
  • 2,692 posts

Posted 04 March 2008 - 11:01 AM

Good evening Bennett,

Assuming there is plenty of condensing capacity available, running the tower with lower reboiler temperatures will result in same degree of separation due to increased relative volatility of key components. Additional care has to be taken, however, in order to supress tower flooding, excessive downcomer backup and - as I have witnessed in one naphtha stabilizer unit - occurance of critical heat flux in reboiler and actually reduced heat input, insufficient to meet process quality targets. Based on these facts, all criteria affecting tower operation should be considered in early design stages.

Almost every application that I had experience with, has some unique and imanent characteristics. Maybe my statement regarding lower operating pressure was too much general, but I think this is the correct approach if all other criteria can be met within reasonable and acceptable limits: CAPEX, turndown, control issues, operability, flexibility of handling different type of feeds etc. Sometimes, increasing operating pressure of the column can be the only way to combat against various operational problems.

This topic is beautiful and definitely one of my favorites. I'd like to recommend you few internet resources, where a lot of quality materials regarding distillation can be found:


Best regards,

#9 mishra.anand72@gmail.com


    Gold Member

  • Members
  • 95 posts

Posted 04 March 2008 - 11:14 AM

Yes, reducing column pressure bring down load on reboiler but increases load on overhead condensor. Check out Cooling water supply temperature and dew point of overhead vapour. Reducing too much pressure will affect throughput of column. If you are really interested in operating column with lower pressure, specify it to design team. They will come out bigger column and overhead condensor.

#10 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 5,490 posts

Posted 04 March 2008 - 11:35 AM


You state: "My position is "strongly held but poorly documented." Experience has taught me to be ready to correct my model and I am prepared to do so in this case--should it be appropriate. Could some good engineer offer some intuition to a chemist?"

I believe you can be strongly documented, if you refer to the Texas Energy Conservation Program Distillation Column Operations document in Andy Sloley's referenced webpage I cited. I realize it is a lengthy document (about 100 pages) - but well worth it because it cover the essence of what you are after: the maximum economical operation of a distillation column with respect to energy savings.

For example, look at section 4-E. "Improving Control of Distillation Columns", where I quote the following:

"At the ISA meeting in Houston, in May 1978, Mr. D.E. Lupfer presented a paper on manipulating the distillation column pressure to increase production and save energy. Operation of columns at the lowest pressure without flooding the column or overloading the condenser, had been practiced by Mr. Lupfer on hundreds of columns. Shinskey (Appendix 7-C), Skrokov (Appendix 7-C), and Fauth and Shinskey (Appendix 7-C) have also discussed the benefits of operating the columns pressure as low as feasible. In the Fauth article, the floating pressure control was part of an advanced control system for a typical gas plant depropanizer. Of the total cost reduction of $1269 per day by using the control system, $345 was attributed to energy savings by the floating pressure control systems.

The floating control systems operation is discussed in the Shinskey and Fauth article. Although the cost of the instrumentation for floating the pressure is low, column temperatures can no longer be used for control because they will vary with pressure. Thus, other control devices such as analyzers, or pressure compensated temperature measurements are required. Shinskey made the following comment in his article on "Control Systems Can Save Energy":

'At first, operators are skeptical of floating-pressure control - they feel more comfortable with constant pressures and temperatures. When its contribution to energy savings is pointed out, they are generally willing to try it. After a brief trial period, they learn that it does not interfere with quality control, and even increases production capacity; soon it becomes accepted. Yet at each installation and with each new application, the concept of floating specifications needs to be sold again.'

Benefits as high as a 30% reduction in energy usage are reported by Shinskey, so the floating pressure control system deserves serious consideration."

As some Forum members will recall, I have always been a very keen admirer of Greg Shinskey - and this is just another reason why. My friend Zauberberg is very correct. This is a beautiful subject as well as a logical, practical, and rewarding topic.

#11 pawan


    Gold Member

  • ChE Plus Subscriber
  • 162 posts

Posted 07 March 2008 - 05:56 AM

Quite interesting topic, in fact it gave me an area where I am going to put it on my blog with some nos later on. I do not agree with this. In my opinion Energy Saving & Energy Requirement are two different things.

In my opinion lowering of pressure has following points.

Consider a fixed feed / overhead & bottom composition column only pressure is changed. U need to maintain same flows & same reflux.

1. Lowering of pressure raises latent heat requirement on TS curve which means you need more energy for boiling the same amount of liquid & turning it into vapors.

2. However this requires low level heat whcih can be catered from hot water or LP in comparison to HP steam So utilization of low grade heat is improved.
This way you can first generate power from HP to LP & then use this LP for distillation, but remember that total LP flow required for reboiler shall be more compared to HP not only by a factor of enthalpy diff in HP & LP but additionally by a difference in latent heat of liquid due to low pressure.

So U need more but still U save.

3. In case of low pressure Vapor compsoition is favorable (higher) than high pressure case in equilibrium to a given liquid composition by Henry / Roults Law. So U need Lesser height but your objective is to achieve same top composition so U can have lesser flows inside the column if it is an older one. This may reduce your LP requirement but net will not go down. However if tower is new it will not be available to you as this margin is eaten up by the designer with even higher capital.

4. Top condenser utility temperature should be much lesser and may result in need of chilled water etc. which is costly affair and again due to higher latent heat U need more. So while U may gain power from HP to LP in case of low pressure U need more power for chilling also.

Well gentleman unfortunately I am going for a meeting so remaining part in next post..........

#12 Bennett Willis

Bennett Willis

    Brand New Member

  • Members
  • 6 posts

Posted 02 April 2008 - 06:58 PM

As I think Pawan noted, he is in disagreement with several other posts in the thread. I was trying to limit the assumptions on this to an oversized column (both reboiler and condenser capacity available along with adequate turn up/down). Is it appropriate to just drop the pressure as much as you can conveniently do? Will you be rewarded for your effort by a drop in the amount of steam consumed? If so, about how many per cent reduction are you talking about?

My original situation was that a collegue had told students that you ran columns under vacuum in order to save significant energy (absolute energy consumption in the column, regardless of the source). It would simply take noticably less energy to run a vacuum column. When I told them that dropping the pressure on the column did not make a lot of difference in the energy consumption, they reported that I was in disagreement with my collegue (but not exactly that politely). I have no particular problem with that, but am trying to find out what the correct answer is--if there is a general answer.

The present preponderance of opinion seems to be that lower pressure does reduce the energy needed (absolute) even as it may complicate vapor velocity (column size) or overhead condensing capability. The way I saw column operating conditions decided (which set everything else) was that you had to boil the bottoms without burning them and condense the overhead with air or cooling tower water if it was at all possible--but I have been in academia (community college) for 10 years and folks may have gotten smarter.

Any way, I'm just looking for other comments.

Bennett Willis

#13 gvdlans


    Gold Member

  • ChE Plus Subscriber
  • 619 posts

Posted 03 April 2008 - 02:35 AM

It has been a few years ago since I was involved in distillation column control schemes, but from what I remember, Shinskey takes advantage of the fact that there is always some fat in the column design, and that your cooling water or air temperature will in general be lower than what was assumed during the design. So with these control schemes pressures are lowered, while at the same time it is ensured that all relevant restraints are met.

When someone proposes that a distillation column should preferably be operated at vacuum in order to save energy, this is a bit different story. This may well be true when looking at utility consumptions, but it comes at the cost of a much larger column. So you may save operational costs, but this is outbalanced by the increased capital investment.

#14 Ali66


    Junior Member

  • ChE Plus Subscriber
  • 19 posts

Posted 04 April 2008 - 09:10 AM

Hello Bennett,

To be specific to your water/ethylene glycol and 90 mm Hg to 250 mm Hg, I think there will be energy savings at the lower pressure, but marginal as the pressure change and thus temperature changes are small. That assumes reflux ratio is kept the same.

The VLE of water-EG widens, becomes more favorable, for separation at lower pressure, so one might say theoretically one can reduce the reflux ratio and obtain same purities. Theoretically this is probably true for water-EG, but again the savings are marginal.

If we do the numbers, I will be surprised if we exceed 5% energy savings when going from 250 mm Hg to 90 mm Hg.


#15 pawan


    Gold Member

  • ChE Plus Subscriber
  • 162 posts

Posted 05 April 2008 - 05:54 AM

Dear Willis

I left my earlier post incomplete. Meanwhile I do not totally disagree with all the forum experts but my view is different from process engineering angle in totality.

Let us consider a fixed feed composition & flow rate with all bottom & top flow rates fixed at 100% purity. & Column is operating at a pressure P.

Now reduce the pressure
So that means essentially all the reflux is 100% top product and withdraw is 100% bottom product. In this case, when you reduce the pressure latent heat of both streams will go up & hence you need more condenser duty for the same top flow rate.

This is nothing to do with the relative volatility, VLE, column design etc. Add losses to this & you will end up with more energy requirement finally.

Also consider following things.

1. In case of low pressure of column you loose pressure energy with respect to your feeding system.

2. In case of low pressure you need higher energy for downstream sections to bring the products at same pressure level. So add pumping energy in both top & bottom products.

3. These two points will vary depending on total system requirement & hence they are not fixed. (Case to case variation will be there).

4. The energy cost should be considered in terms of equivalent fossil fuels not just steam or power in such comparisons. This is the most critical issue & is always my focus.

5. In case of vacuum systems you need to consider the operating cost of vacuum system as well, higher the vacuum higher the steam consumption in ejectors. But that is relatively less in comparison to columns if throughputs are higher.

6. U may need chilling cost if you drop it below CW temperature. Include running cost of chilling unit.

Any counter discussion for this interesting topic.

#16 djack77494


    Gold Member

  • ChE Plus Subscriber
  • 1,282 posts

Posted 07 April 2008 - 01:32 PM

You have generated a lot of interesting opinions about a very interesting topic, and you've gotten many of the most experienced and qualified forum experts to weigh in. Congradulations (and thank you) for that.

My personal opinion is that it is dangerous to reach for generalizations such as "lower pressure separations result in energy savings". I'm sure you are aware of this. An alternative approach to discussing this topic might be to review aspects of separation that would tend towards higher or lower pressure operation. Even this is a "tall order", and I'll just "scratch the surface" of this topic here.

1) Easiest to discuss would be situations where pressure/temperature is THE key consideration in effecting a separation. Where a heat sensitive material is to be fractionated or distilled, you are forced to select a pressure where the required temperatures will not degrade your product. Many vacuum separations come about due to this cause.
2) Higher pressure separations have a number of advantages including:
a - Gas phase is denser so (usually) there is less risk of flooding and the ability to process more material.
b - Cheaper condensing utility can be used (e.g. cooling water rather than refrigerant).
c - Reduces need for downstream gas compression and bottoms pumping.
(Intuitively, cold processes would tend to favor higher pressure separations.)
3) Lower pressure separations have a number of advantages including:
a - Better separation between components due to improved differences in vapor pressures.
b - Lower temperature operation.
c - Cheaper heat utility can be used (e.g. lower pressure steam).

When you put these factors together, it remains difficult to make generalizations. The cost of a smaller column that could be used for a high pressure separation is offset by the higher design pressure and temperature. Choice of materials of construction might be affected. So the net effect on capital costs could go either way. In a similar manner, depending on a variety of factors, perhaps especially on the utility situation in your plant, the operating costs also may be favored by either higher or lower pressure operation.

In my opinion it is a very valuable skill to be aware of the tradeoffs involved and to have some "feel" for the economic impact of the alternative options. An experienced engineer with good judgement can often make "from the hip" assessments of key process parameters like this that will match or exceed the quality of result that can be gotten from a detailed study by less experienced engineers. My recommendation is to strive to understand the process. (Did I say one should avoid generalizations?)
Good luck,

Similar Topics