15 replies to this topic

[sjtay]

Hi all,

I am a final year student in australia and am trying to complete my design project. I am tasked with designing the equipment for the Iron Sponge (oxide) process for the removal of hydrogen sulphide from natural gas.

I have come to understand that this is a fixed bed adsorber column design but i cannot find much literature on how to design it.

My university lecture notes from a previous subject (process equipment design) only tells us how to design packed bed absorber columns.

However, my iron oxide bed is solid and stationary while the acid gas flows from top to bottom with a reaction occurring as it flows.

Does anyone have a general document or website i can refer to that has the design instructions for a fixed bed adsorber column? Such as how to calculate how much we need, column heights, column diameters and so on. I have the flowrates and mass fractions entering and exiting.

Any help would be great Thanks.

Hello Sjtas,
iI think this title (you can in internet found) useful "design review of absorbers used for gaseous pollutants".

Nabeel Hayder

[breizh]

http://www.google.fr...iw=1125&bih=592
Hi Sjtay,
Did you try Google?

Breizh

Edited by breizh, 02 October 2012 - 06:16 AM.

[bikrom]

Aren't packed and fixed beds one and the same?

[Art Montemayor]

Sjtas:

Why haven’t you at least spent some time researching for literature on this very old process? You obviously have use of the Internet and there is tons of information available on this basic process that uses a chemical reaction to eliminate hydrogen sulfide from a carrier gas. Please refer to the attached file that can easily be found at: http://lib.dr.iastat...291&context=etd

I have even highlighted the area of interest you seek. It took me 5 minutes to find this information. I hope you take this critique as a constructive one and a learning lesson. (And I note that while I have been writing this, Breizh has also arrived at the same point.)

Contrary to the author of this thesis, I do not qualify this process as “adsorption”. It involves nothing more than a basic, first-year, chemistry equation. Adsorption – as I was taught – is a Chemical Engineering Unit Operation. It does not involve a chemical reaction and mainly depends on surface van der Waal forces to bind the desired component until the base adsorbent is regenerated. In the sponge iron process the base reactant – iron - is not regenerated. Note that the author of the thesis is a civil engineer – typically with only one semester (8 hours) of chemistry courses. My point here is that I would deal with this process as a Unit Process (which involves a reaction) and not as an adsorption process. That being the case, what we normally have done in industry is to simply employ an empirical Space Velocity to design the fixed reactant bed.

Attached Files

•  BioGas Treated with Iron Sponge.pdf   5.37MB   183 downloads

[ankur2061]

sjtay,

I suggest you go through the excellent discussion on the subject matter of desulfurization using Zinc Oxide which uses a similar reaction method as iron sponge does for removal of trace quantities of H2S at the link below:

http://www.cheresour...hot-zinc-oxide/

Regards,
Ankur.

[sjtay]

Thank you for all the replies and advice. However, i have tried googling and searching the internet. I am looking for a way to design the major equipment of this process. Steps that involve finding any equilibrium data, breakthrough curves, column diameter, column heights, pressure drops, inlet and outlet nozzles, and so on.

I have notes on how to design a packed absorber column but that involves a gas and liquid flowing counter currently. then the VLE data must be determined in order to plot operating lines and then figure out the number of ideal stages etc.

However, this involves a solid bed and a gas flowing downwards. Its not that I haven't looked, but I am running short of time and thought if someone could help, that'd be great.

[Art Montemayor]

Sjtay:

I would not go into any attempts to try to model or calculate the size of the required bed using equilibrium data, breakthrough curves. I recommend you follow the same philosophy I show in my uploaded Zinc Oxide reactor file in the thread referred to by Ankur. It really is that simple as shown in my sketch and calculations.

You design for two vessels in series, that are valved in order to “stagger” their stage of operation. In other words, you monitor the H₂S content in-between the two vessels and when it appears that the H₂S is about to break through the first, you shut down the first vessel (continuing with the second on stream), replenish it with fresh reactant and put it into service as the second vessel in series. That way, you don’t stop production due to the need to change out spent reactant.

You design the beds based on the chemical reaction and a decent superficial velocity (0.6 meter/min is suggested in the thesis reference). Knowing the stoichiometry of the reaction and design velocity through the beds, you can calculate the estimated time for the consumption of the bed. Or you can design for the bed consumption time. Knowing the design velocity, you can calculate the design bed diameter and with that and the reactant quantity you calculate the bed height and the resulting pressure drops. Your design is essentially straight-forward and you should have only one problem: using your common sense to guide you to what you need.

[sjtay]

Thank you very much for your help.

One more questions: How do i convert kg/hr to MMSCFD?

What i have so far is my mass fractions of my gas mixture, flowrate in kg/hr.
My process operates at 45 C and 180 kPa. How do i convert kg/hr to MMSCFD?

[breizh]

SItay ,

One is related to the other by the density >>>> You need first of all convert your flow rate (MMSCFD million standard cubic foot per day to let say Standard ( 15C , 1bar ) cubic meter per hour then to Cubic meter (45 C, 180 kpa) per hour .

let you use the perfect law gas to convert Standard to Actual .

For the last part you need to calculate the density , for this part you need the composition of the gas to calculate the molecular weight and again using the perfect law gas , you will access to the density = Mw/V = Mw*P/(R*T ).
If you don't want to do hand calculation , you can download very useful tool like Uconeer to perform the calculation.


Breizh

[sjtay]

My flow rate is in kg/hr. 89640 kg/hr to be exact. I want to convert to MMSCFD not the other way around.

My mass composition is:
CO2 - 0.9750
H2S - 0.0001
CH4 - 0.0019
H2O - 0.0228
C2H6 - 0.0002

How do i get the overall density of the gas? Is there a way to calculate?

From there, how do i start converting to MMSCFD?

[sjtay]

I am following the highlighted part of this document. The only problem is my values are in different units.

My values:
Flowrate = 89640 kg/hr
Op. Pressure = 180 kPa
Op Temp = 45 C
Z = 0.9912

But the document uses:
Flowrate in MMSCFD
Op Pressure in psig
Op Temp in Farenheit

I've tried using various online sources to convert my flowrate into MMSCFD and then following the document attached, to find the actual gas Flowrate Qa, i got 18078.55. But i dont know what unit this is in. Is it in MMCFD rather than MMSCFD?

Part 2 asks me to find the internal diameter and internal cross sectional area and a formula is given. But how do i get the cross sectional area? On the right there is a table saying for iron sponge, the cross sectional area is 0.1Qa, which gives me 1807.85 square ft (167.95 m^2)

Using this area to find the ID, i get an ID of 47.98 ft or 14.62 m. Is this value too large for a column diameter?

Please refer to the document attached.

Attached Files

•  GasSweeteningWorkshop_2igf_June2008.pdf   870.99KB   100 downloads

[Art Montemayor]

Sjtay:

I have worked out this type of basic, chemical engineering calculation many times in our Student Forum in the past. You can access this information (and spreadsheet calculations) using our SEARCH engine. You can also do it another way, using the calculations suggested by Ankur in: http://www.cheresour...ard-conditions/

I personally always resort to converting so-called “standard” volume condition using the value 379.49 Scf/lb mol. That is recognized as the molecular volume and it doesn’t change. You can find this value in the GPSA (I won’t waste time discussing the GPSA, because I believe it is world-known and you can easily research it on this Forum and Google).

It is well-known basic chemical engineering that one lb-mol of any gas will occupy 379.49 ft³ at 60 ^oF and 14.696 psia (which are the GPSA definitions for Scf). MM simply means MILLION (M = thousands). That is the USA conventional engineering units used in the USA and spread all over the world in the last 60 years or so.

You need to know the molecular weight of the gas handled. Knowing the pounds/day of gas handled, you convert this into lb mols of gas per day by dividing the lb mass by the molecular weight. The resulting lb mols of gas/day is multiplied by the molecular volume (379.49) and the result is MM Scfd of gas. It is that simple. The key to this calculation is the molecular volume - which is based on Avogadro's Number.

If you need to convert mass flow rate into Scfd, you do the reverse. I am not going to go into converting USA conventional units into metric units. That, I believe, is well below the level of education required for this Forum. For that, use the excellent and convenient program designed and distributed by one of our members, called UConeer. Our member Breizh recommends it also. You can also find reference to this program in our Forums, using SEARCH.

I hope this information allows you to continue onwards to more complex problems.

[breizh]

Hi,

P = 1 bar, 15 C >>>> Z = 0.995 & d = 1.82 kg/m³
P = 1.8 bar, 45 C >>>> Z = 0.9932 & d = 2.97 kg/m³

Hope this helps
Breizh

[sjtay]

To Art Montemayor:

Thanks so much. That helps a lot. But just to clarify some things:

Once i have multiplied lb mols/day with the molecular volume 379.49 Scf/lbmol, do i divide it by one million to get MMSCFD?

[Art Montemayor]

SJtay:

Since someday – hopefully - you will be a chemical engineer, my strongest advice for you is to ALWAYS USE ACCURATE AND DETAILED UNITS in your calculations. For example note the following:

(100 lb mols/day) (379.49 Scf/lb mol) = 37,949 Scf/day

(37,949 Scf/day) (MM/1,000,000) = 0.03795 MM Scf/day

This proves your suspicion is correct.