12 replies to this topic

[hoyoku]

Can anyone tell me how heat transfer for cooling or heating works in a trickle bed reactor? For example, a batch reactor can have a cooling jacket to maintain the temperature of the reactor.. But trickle bed reactor, what is the cooling/heating device used? Cooling jacket? Or something else. Thanks for helping. Urgent!

[Art Montemayor]

Hoyoku:

This is not only for your benefit, but for all those other ill-organized ChE students who wait for the ultimate, 11^th hour to start work on their assignments and then proceed to do the last-minute task under an “URGENT” atmosphere. All this type of bad organization and procrastination on assignments does is create bad work and mistakes stemming from bad decisions. Let me give you an example:

You have made a serious and totally wrong statement by saying “a batch reactor can have a cooling jacket to maintain the temperature of the reactor”. There is no cooling jacket on a reactor that I’ve seen or heard about to date that could realistically control the temperature of the reaction taking place inside the kettle. This type of thinking stems from not asking the right questions or reading the wrong books – or more often than not, listening to the wrong professors who have never operated or controlled a batch reactor.

A jacket is installed on a kettle for the simple reason that it is nothing more than a “token” attempt to use available surface area to either cool or heat up a reactor. It is a “token” amount of heat transfer area simply because it is limited due to the fact that the size of the kettle is not dependent on the heat transfer as it is on the kinetics of the reaction(s) taking place and on the size of the batches that are decided upon by economics or production needs. The most useful work that a jacket on a kettle type reactor does is probably heat up the initial reactants as they are poured into the kettle – and this is for a liquid-based, stirred reaction. When you are dealing with a fixed catalyst bed type of reactor – as I believe you have – you don’t have the ability to stir the reactants nor do you have a good manner to transfer heat – except by direct contact between the reactants or the carrier fluids. You have not fully described what you mean by a “trickle bed” reactor, but what I’m accustomed to calling that type of reactor is a fixed catalyst bed that is fed liquid reactants that combine with a recirculated gaseous phase to form either a condensable vapor product or a liquid product within the reactor. Under this type of scenario, what is commonly used is direct, adiabatic cooling with the carrier gas (which can also be part of the reactants).

If you are employing a tubular reactor with the tubes packed with catalyst, then you can have realistic heat transfer control by circulating heat transfer oil around the tubes. What you essentially have, in that case, is a vertical heat exchanger with the tubes packed with catalyst. The reaction takes place inside the packed tubes and the shellside has recirculated oil to remove exothermic heat or to supply heat for an endotherm. Is that what you have (or are proposing)?

For example, I have designed and produced Furfuryl Alcohol (FA) in such a fixed bed, tubular reactor by injecting a vapor mixture of Furfural and Hydrogen through the packed tubes. I controlled the bed temperature by circulating oil in the shellside of the vertical reactor. The product was a vapor mixture of FA and excess Hydrogen. The Hydrogen served as a carrier gas and helped to control the reaction and take it to completion. An external condenser condensed the FA and the excess Hydrogen was recirculated.

Be specific in explaining what you are proposing – do not assume we all know the acronyms and titles that you give your equipment. Describe everything in detail to enable all of us to have an accurate idea of what it is you are proposing. Better yet, sketch out your process on an Excel workbook and attach it to your post. An engineering sketch is far better in explaining what it is you are proposing. Once we have an accurate description of what it is you are proposing, we can better help you.

Above all, give up on designing a reactor jacket to be the source of your controlled heat transfer. Any knowledgeable professor will laugh his tail off when he sees this as a proposed solution to control a batch reactor. It is never done this way in industry because you cannot rely on it and it is not practical. You must employ other means, and when you submit an accurate idea of what you propose, together with identification of the reactants and products, we will be in a better position to assist you.

Await your reply.

[hoyoku]

Thanks for your advice... My process is a high pressure hydrogenation between methyl ester [CH3(CH2)10-C-OCH3] and hydrogen to form lauryl alcohol [CH3(CH2)10C-OH] and methanol. The process is carried out at around 270 celcius and 200 bar. There are two types of reactor that can be used. One is a fix-bed reactor and one is a trickle bed. The fix bed reactor employs a gas phase hydrogenation, whereas a trickle bed has liquid methyl ester and hydrogen gas coming in. I've attached a picture of a trickle bed reactor for your reference. I've also attached my preliminary pfd for your reference. Basically I have 3 main parts in my plant, because I'm producing Sodium Lauryl Sulfate [CH3(CH2)10C-OSO3Na] from Lauric acid [CH3(CH2)10COOH]. The first part of my plant produces Methyl Ester as a raw material for my hydrogenation plant. So the trickle bed reactor I'm talking about is R-201 in the pfd. What you have said is correct about trickle bed reactor:

"I’m accustomed to calling that type of reactor is a fixed catalyst bed that is fed liquid reactants that combine with a recirculated gaseous phase to form either a condensable vapor product or a liquid product within the reactor. Under this type of scenario, what is commonly used is direct, adiabatic cooling with the carrier gas (which can also be part of the reactants)."

But what I don't understand is the part where u mentioned the direct, adiabatic cooling with carrier gas. Can you further explain this to me?

And another question I would like to asked is if I were to use a tubular reactor like your FA reactor, what is the practical way of vaporizing all your alcohol before it enters the reactor? I tried running the simulation with HYSYS, but I can't get a 100% vapor alcohol at the temperature (270 celcius), which is my reaction temperature. Is there any realistic way of solving this problem?

I still have 5 more weeks before I submit my mass and energy balance, so I wanna try my best to get things right before I enter the presentation room. Your advice is very much appreciated. Thanks a lot. Have a nice day! Any extra information that you need, pls let me know.


Joseph

Attached Files

•  Trickle_Bed.png   60.35KB   57 downloads
•  PFD_hoyoku.jpg   124.45KB   51 downloads

[hoyoku]

And one more thing, does it mean that even for continuous stirred tank reactors, a cooling jacket can't realistically control the temperature? Then how do they control the temperature for CSTR then?? Most of my assignment questions involving stirred tanks employs cooling jacket for temperature control. We even design PID controller for processes with cooling jackets. Appreciate your reply.

Joseph

[hoyoku]

Extra information concerning my PFD attached in previous reply. The first part of the plant is an esterification plant, second part is hydrogenation, and third is a sulfation plant.

CH3(CH2)10-COOH + CH3OH = CH3(CH2)10-COCH3 (First plant)

CH3(CH2)10-COCH3 + H2 = CH3(CH2)10-COH + CH3OH (Second plant)

CH3(CH2)10-COH + SO3 = CH3(CH2)10-COSO3H (Third plant)

CH3(CH2)10-COSO3H + NaOH = CH3(CH2)10-COSO3Na + H2O (Third plant)

[Art Montemayor]

Joseph:

I think there is still time to structure a decent and successful mass and energy balance of a credible and practical process.

First, concentrate on the basis of your reaction: the kinetics involved and required as well as the required phase under which the reaction takes place. These principles set the basis for the type of reactor you require or is the most practical for the reaction. If you require to carry out the reaction of your liquid + gas reactants in the liquid phase, then you probably have a simple and easy reaction. However, if your reaction has to take place in the vapor phase, then you have more work in front of you and the reactor is more complicated – as well as the subsequent condensation and separation of the reaction products. You are the one in charge of setting the reaction parameters, so you have to tell us what phase you are going to run the reaction in. You have not mentioned anything about the phase of the reaction – only the types of reactors you are thinking about. The kinetics and the physical properties will determine the phase under which you will run the reaction.

In a Continuous-Stirred Tank Reactor (CSTR), one or more fluid reagents are introduced into a tank reactor equipped with an impeller while the reactor effluent is removed. The impeller stirs the reagents (one of which has to obviously be a liquid) to ensure proper mixing. If your kinetics call for the reaction to be carried out under a liquid phase, you can still employ a fixed, catalytic bed type of reactor, but you obviously cannot stir the reactants. You now have to depend on critical distribution and a high Reynolds Number to obtain turbulent flow and good mixing. This is the first step in designing your process: you must decide what type of reactor you require to succeed in carrying out the required rates of reaction and subsequent separation in a safe, practical, and economic manner. This is your decision.

In a fixed bed type of reactor you have adiabatic cooling taking place because you feed excess gaseous reactant (Hydrogen, in your case) and this gas has a good heat capacity value that enables it to carry off heat with it as it is recirculated around the reactor system. You will instantly become aware of this if you are paying careful attention when you do your heat and mass balance around the reactor. The excess Hydrogen that you have to feed – not only to ensure complete reaction, but also to stimulate turbulence – will also serve as a “heat sink” that carries away some of the exotherm generated in the reactor. Your external product (& excess Hydrogen) cooler then serves as heat control device for the reaction.

If you can’t vaporize the liquid reactants at the pressure you are running the reaction, then you are stuck with running a liquid-phase reaction. It’s that simple.

Kettle reactors cannot be temperature controlled by only depending on their inherent jackets. This is totally wrong thinking. You have to install one of two basic methods to ensure the critical control of a kettle reactor:

1. You install internal cooling/heating coils that employ the convective action of your agitator(s); or,

2. You install an external heat exchanger which employs a “pump-around” circuit with control valves to temper the amount of cooling or heating you require. A pump-around circuit is nothing more than a centrifugal pump that pumps the contents of the kettle out and through an external heater/cooler and then returns the fluid back to the kettle.

I hope this helps you out.

[hoyoku]

Thanks for your advice. By kinetics do you mean the rate equation, rate constant, activation energy and stuff of the reaction? In my first plant, which is for esterification. The reaction between lauric acid and methanol is being carried out at 180 celcius and 1 atm in a CSTR. So I proposed to heat the lauric acid to the reaction temperature, but methanol will go in at room temperature because I don't want the methanol to be in vapor phase when it goes in. So does that mean that, I need to have a heating coil or something in the CSTR for maintaining the temperature around 180 celcius? Do they have heating coil in CSTR? I've never seen a CSTR before. And at 180 celcius, I suppose methanol will vaporise in the CSTR, so I'm proposing a reflux of methanol back to the reactor.

For gas-liq separator such as the gas-liq cyclone, is it possible to assume a 100% separation? In my third plant, I have SO3, Air, CH3(CH2)10C-OH and CH3(CH2)10-COSO3H coming out from my sulfation reactor. SO3 and air is gaseous phase where as the lauric acid and lauryl alcohol sulfuric acid is in liquid phase. And I want to separate the SO3-air from the liquid products. Is it reasonable to assume 100% separation in the phase separator where SO3-air all goes up and all liquid goes down? What is the most important thing in dealing with gas-liquid mixture coming out from the reactor?

Thanks!

Joseph

[Art Montemayor]

Joseph:

You’re not helping me help you when you persist in avoiding discussion of the details of your reactions, as you are proposing them. When I say “kinetics”, I mean the whole ball of wax – all the details that make for a practical, controlled and efficient reaction to take place. You are failing to tell us the specific temperatures, pressures and phases that you are proposing. Also, now you are discussing more reactors. Keep in mind that you have all the details of what you intend to do in your mind and if you don’t communicate them clearly, efficiently, and accurately, we can’t keep pace nor remember all that you have said or proposed. Attack only one reactor at a time. You are only confusing your readers (& yourself) if you keep introducing other reactors and conditions in the same paragraph or thread. Select a reactor you want to design. Identify the reaction taking place by defining the kinetics involved – such as the temperature, pressure and phase of the feed streams, the reactor, and the product streams.

This thread started with the high pressure hydrogenation of methyl ester [CH₃(CH₂)₁₀-C-OCH₃] to form lauryl alcohol [CH₃(CH₂)₁₀C-OH] and methanol. Now, you’ve introduced esterification and sulfonation. Also, bear in mind that chemical equations define what is being reacted – and that’s good. However, your chemical equations:

CH3(CH2)10-COOH + CH3OH = CH3(CH2)10-COCH3 (First plant)

CH3(CH2)10-COCH3 + H2 = CH3(CH2)10-COH + CH3OH (Second plant)

CH3(CH2)10-COH + SO3 = CH3(CH2)10-COSO3H (Third plant)

CH3(CH2)10-COSO3H + NaOH = CH3(CH2)10-COSO3Na + H2O (Third plant

don’t tell us the temperature, the pressure, the catalyst, the form of the catalyst, nor the phases of the reactants, the reaction, and the products. You know the kinetics (or you should) so only you can give us that information. It is a waste of time to discuss types of reactors if we don’t know the form of catalyst required or all the phases involved. The correct and optimum type of reactor depends on the phases you need to handle, the type and form of catalyst, the pressure and the temperature. You want to avoid as much as possible any handling of 2-phases flow – especially through a catalyst bed. The reason(s) for this is common sense. If you feed a 2-phase mixture into a fixed catalyst be, you are presuming that the reaction will take place whether it is in the liquid state or in the vapor state. This is usually not the case. If the reaction takes place in the vapor state, then you have to furnish the latent heat of vaporization for the liquid reactant INSIDE the reactor – which is out of your physical control. You don’t want to have to design for this type of difficult (if not impossible) situation. It is far simpler to vaporize all liquids outside the reactor and feed only vapors – if the reaction is to take place in the vapor phase. This immediately eliminates any kettle or CSTR type of reactor and puts the selection in the area of a fixed catalyst bed vessel – whether there is one, monolith bed or there are packed tubes. The packed tube type is selected when you need rapid and critical temperature control of the reaction, while it is taking place.

It is rare that you have a steady-state reaction where you have 2-phase products coming out of the reactor constantly. Normally, when you hydrogenate for fine quality products, you employ batch, kettle type reactors that are stirred and cooled with internal coils – although I’ve also used external, recirculated coolers when there has been a potential hazard of a run-away reaction. Examples of this are the hydrogenation of fats and oils in the food industry.

If you are hydrogenating methyl ester:

CH₃(CH₂)₁₀-COCH₃ + H₂ = CH₃(CH₂)₁₀-COH + CH₃OH

In a batch reactor, that’s one thing. However, if you need to have a continuous, steady state reactor that is another situation. For a continuous reactor, you will have to address the fact that you can’t achieve a definite % completion of the reaction in just one pass of the reactants. You will have to recycle to control the conversion rate and vent off the excess Hydrogen to a recompression system that recycles the H₂. A batch system, of course, is much simpler to design and operate.

So, you have to define to us in detail just precisely what it is that you propose. Then we can all jump in to help you if we can.

[hoyoku]

Let me go step by step. As I've mentioned, I'm design a plant to produce sodium lauryl sulfate [CH3(CH2)10-COSO3Na] from Lauric Acid [CH3(CH2)10-COOH]. Therefore, the first process involves:

1) Esterification of Lauric Acid and Methanol:

CH3(CH2)10-COOH + CH3OH = CH3(CH2)10-COCH3 + H2O

Feed Lauric Acid: 180 C, 1 atm , liquid phase
Feed Methanol: 30 C, 1 atm, liquid phase
Product Methyl Laureate (Ester) : 180 C, liquid phase
By product-H2O: 180 C, vapor phase
Unconverted Lauric Acid in product: 180 C, liquid phase
Unconverted Methanol: 180 C, Vapor phase
Catalyst: Conc. sulfuric acid (liquid)

Feed methanol to lauric acid ratio: 3-4 mols methanol per mole of lauric acid.
Conversion: 90%
Reaction Time: 30 minutes

The esterification reaction is to be carried out at 180 degree celcius and 1 atm. The catalyst employed is concentrated sulfuric acid (liquid). And I'm proposing a CSTR for the reaction. For your information, the boiling point of methanol is around 65 degree celcius, and obviously at the reaction temperature of 180 degree celcius, some methanol will vaporise. The lauric acid (liquid phase) that enters the CSTR will be preheated by a heat exchanger to 180 degree Celcius, which is the reaction temperature. And I'm proposing that feed Methanol be left at the room temperature, since I do not want methanol to enter the CSTR in vapor form. So in the CSTR, at 180 degree celcius, lauric acid at 180 degree celcius will remain as liquid as its boiling pt is far above 180 C. But my problem is, altho methanol enters as liquid at room temperature, once it enters the CSTR and is mixed with lauric acid at 180 C, I suppose that methanol will vaporise. The H2O that forms will also be vaporised at 180 degree C and 1 atm. So I'm proposing to channel the vapor phase which constitutes of mostly methanol and H2O to a distillation column which separates methanol and water, and the methanol that comes out from the top stream of the distillation column will be condensed from vapor to liquid by a cooler and fed back to the reactor. My question would be:

1) I certainly want only a single phase (liquid) to occur in the reactor, but I cannot avoid having methanol and water to vaporise since the reaction temperature for esterification from literature is usually 180 C. Since I can't avoid having methanol and water to vaporise in the reactor due to the reaction temperature of 180C, so is this process reasonable for practical purposes?
2) In order to maintain my CSTR temperature at 180C, does it mean I need to have a heating coil or something for temperature control?
3) If there is any better suggestion or modifications that you could propose, pls do.

I'll stop at this first process first as I do not wish to confuse the readers. Thanks for your advice.

Joseph

[Art Montemayor]

Joseph:

You have failed to tell us that this is a Batch Reactor – one in which you will initially fill in the base Lauric Acid together with the sulfuric acid catalyst. You will introduce the methanol reactant at the rate that the reactor can generate a safe exotherm that you can control with cooling fluid. I am assuming that the esterification is an exotherm – which you also failed to advise.

If the above is the case, then refer to my attached workbook where I have sketched out a simple batch reactor set up that I foresee for this type of reaction. Note that you must do your process design work first. You must determine the pressure that the reactor will work at. The pressure in the reactor at any time while it is working is the vapor pressure of the fluids inside the reactor. As you consume more and more methanol, you will be producing proportionately more water. The highest possible pressure inside the reactor (at the reaction temperature of 180 ^oC) seems to be the corresponding vapor pressure of pure methanol, which is 27.158 Bara (407 psia). You would be wise to design for a pressure in excess of that in order to have a margin of safety. I would expect a design pressure of around 500 psia.

You can control the reaction rate by cooling with the internal helical coil and/or with the reflux you obtain from the total condenser over the reactor. You must study and be able to predict the effect that the by-product of water has on the reactants as well as on the product Methyl Laureate. I’m guessing that the water will only figure in as a bothersome impurity that has to be tolerated because you can’t vent it while its vapors are mixed with the methanol vapors that you either need or can’t vent to atmosphere until the reaction is essentially complete.

As I see it, this is a pretty simple reaction and reactor controls are direct and flexible.

Do not fail to do your process homework on the reaction system. You must be able to predict what is going on during the reaction - temperature, pressure, and phase-wise.

You, also, can use graphic sketches to show what you propose – just as I have done. Communicating correctly and accurately is part of engineering and forms part of the answer to your problem. Knowing how to solve a problem is only a part of the answer. You must also be ready and capable of explaining correctly what you propose and how you intend to carry it out. Otherwise, you wil find that others will not be able to help you effectively.

 Reactor_Selection.xls   43.5KB   136 downloads

[hoyoku]

Can I propose a continuous process as in the attachment? My supervisor seems doesn't really fancy the idea of batch reactor, although my initial proposal is a batch reactor for the esterification process. He suggested that I use a CSTR, for continuous process. That's why I had to come out with this process flow. Please see if this is a reasonable process. Thanks.

Attached Files

•  My_proposal.xls   70KB   71 downloads

[Art Montemayor]

Joseph:

I don't know what role or authority figure your supervisor plays. If your supervisor writes the rules, then do what he/she says. If your supervisor is there to play a "Devil's Advocate", then challenge the proposal to make the reactor a continuous model. A continuous reactor is much more expensive, complicated, and sophisticated; it must be justified on a present value method. Has this been done? I doubt it. But if your supervisor is writing the rules as you go along, then you must comply - or as we used to say when I was in your place: "Cooperate and graduate".

I can't even make a guess or an opinion, since you haven't even furnished us with flow rates or a completed heat and mass balance. Therefore, we don't even have an idea of the scale of this proposal. Do you grasp now what I've been stressing about the importance of communications?

You must make a decision or take a stand.

[hoyoku]

Thanks Sir..

Once I discuss with my groupmates and have come out with more detail of my design, I will let you know again. Cause now everything is just in the "thought" phase. As you know, different lecturers often have different ideas about design. And most of them think they are right, which in many cases they are. But as a student, sometimes it makes me hard to stand in the middle, having to agree to different opinions from different lecturers. My supervisor says that its awkward to have a batch system (esterification) being joined to a continuous system (hydrogenation). So that's why he said use a CSTR. I will work on the design first, then I will come for your advice again. Thanks a lot for your practical help!