It doesn't make any difference how you define SV as long as you define it and enable "real" information to be extracted from the SV by identifying all of the conditions associated with the calculation of it..  It is a very rough measure of capacity or size of a catalytic reactor so we can get a feel for the relative size of the system or the approximate contact time or turnovers.  It is probably most useful if we define it as the standard volume of gas divided by the reactor volume but other definitions use the actual gas or liquid (liquid is going to be about the same whether actual or ideal conditions) volume.  If the reactor is feeding both gas and liquid then there are two different space velocites.  If someone tells me the SV is 1,000 I get a feel for whether the reactors are going to be big and expensive with a high overall cost of catalyst than if it is 10,000 where the relative catalyst and reactor cost would be 10 times less per unit of product (assuming the same conversion of feed).

Likewise, if you want to correct out the inerts and define the SV on reacting moieties only, you are free to do so if you make it clear that is the case.

If you are going to model a reactor you will never explicitly need to calculate space velocity.  The model would only be concerned with the physical and transport properties and flow dynamics around the catalyst particle and concomitant heat transfer effects. The SV is more of an overall macro property.

Regards,

Dennis Leppin, P.E.

Institute Fellow, Gas Technology Institute

Before answering , what do you want the SV data for? Are you wrinting a report or a patent? Are you reading a paper?

Are you in the US or US-influenced country?

It doesn't make any difference how you define SV as long as you define it and enable "real" information to be extracted from the SV by identifying all of the conditions associated with the calculation of it..  It is a very rough measure of capacity or size of a catalytic reactor so we can get a feel for the relative size of the system or the approximate contact time or turnovers.  It is probably most useful if we define it as the standard volume of gas divided by the reactor volume but other definitions use the actual gas or liquid (liquid is going to be about the same whether actual or ideal conditions) volume.  If the reactor is feeding both gas and liquid then there are two different space velocites.  If someone tells me the SV is 1,000 I get a feel for whether the reactors are going to be big and expensive with a high overall cost of catalyst than if it is 10,000 where the relative catalyst and reactor cost would be 10 times less per unit of product (assuming the same conversion of feed).

Likewise, if you want to correct out the inerts and define the SV on reacting moieties only, you are free to do so if you make it clear that is the case.

If you are going to model a reactor you will never explicitly need to calculate space velocity.  The model would only be concerned with the physical and transport properties and flow dynamics around the catalyst particle and concomitant heat transfer effects. The SV is more of an overall macro property.

Regards,

Dennis Leppin, P.E.

Institute Fellow, Gas Technology Institute

Dear colleague,

If you need to report the SV for a continuous reactor, I suggest that you read the paper by S. M. Walas (Chemical Reactor Data, Chem. Eng., Oct. 18 issue, 1985). You will find the proper SV to use. I am sure that will answer all your questions. The only doubt that can arise is how to use SV if you use metric or SI units, because the paper is US users. But I am sure you will solve the problem.

The paper lists  SV data for 102 referenced industrial reactors, so you will see how SV are expressed. Residence time is only used for batch reactors in practice.

If you teach reactor design I suggest that you recommend the Walas paper to your students. 

I hope this information is helpful.

Best regards,

Francesc Recasens, UPC, Barcelona, Spain

Hello F. Recasens

I am a Chemical Engineering student in Colombia. I need some help with SV for a continous reactor. I looked fot the Walas's paper in the Internet, but I couldn´t find it!. Could you please send it to me if you have it in PDF ?

Thanks