top fired vs side fired reformer for ammonia

[Das Raj]

Hi all, Linde offers top fired reformer for its ammonia technology vs side-fired which is offered by other tech suppliers. I wanted to know what are thre advantages of this design, any thoughts? any links or articles on the comparisons will help.

also is S/C ratio still around 2.7 to 3 for the top-fired reformer? what is the exit temperature?

[Edmondson Bob]

Hi, Experience with both options is largely quite good. Uhde, KBR and Linde have good top fired designs with differences largely being in the outlet header design. Temperature for non-Purifier plants is in the 800 deg C range with S/C in the 2.8 to 3.0 range. Side fired plants also work well with Fluor and CF Braun having convection banks above and Haldor Topsoe, at graded. At least to me, advantages and disadvantages are minor.

[raj da]

Thanks Bedmonds.

800 C seema kinda low. I have seen usually 850-900C at exit of reformers. what is the pressure of these top-fired reformers?especially the ones made by Linde an dused for hydrogen/ammonbia etc? is it around 30bar

[Edmondson Bob]

Hi Daraj,
For ammonia primary reformer outlet ranges from 680 deg C(for Purifier plants) and as high as 820 deg C for conventional plants. After air addition to secondary reformer its outlet tends to be in the 980-1000 deg C range which is a good fit to:
keep down Al2O3 volatility
keep down Ni volatility
prevent DNB on waste heat boiler tubes at the boiler inlet
If 900 deg C exit the primary reformer secondary outlet is 1210 deg which causes problems for the three items above ns CH4 only drops from 0.5% to 0.04% so little is gained despite putting a lot more fuels to the reformer. For pressure the larger plants that I have been involved with have ranged from 30 bar to 45 bar. Linde are in that range as well but they have lots of expertise in the world of H2 plants as well and have good experience with PSA units for CO2 removal which may be applicable for plants not producing urea.

[raj da]

Bedmonds, I guess I was not clear. I am talking about processes which have ONLY SMR, no secondary reformer. In order to have a low methane slip I guess you need to have higher exit temperature. Can you confirm is 28-30 bara and say 860C is good enough to get a 4-6% methane slip?

also in isothermal water gas shift reactors what is the maximum CO conversion that can be attained?

[Edmondson Bob]

Hi,
Understood now -at 30 bar and 3.4 S/C you can have 5% CH4 assuming 5 deg C Approach to Equilibrium. For the shift reactor where Linde use a single tube cooled converter with heat rejected to steam generation I can’t recall the steam pressure that was being raised and that would determine the operating temperature so I can’t really hazard a good guess. It’s complicated because you would not want to have condensation on the outer surface of the tubes and dewpoint would be in the 184 deg C range for this example so to have some margin steam pressure being generated would need to be, as a minimum, around 15 bar. Just rough figures I think that means an outlet around 240 deg C at a reasonable total tube OD area. Perhaps if you contacted the Incitec Phosphate Hill plant (Queensland, Australia) they would be willing to share their situation but it may be that they are prohibited by confidentiality. An interesting subject.

[Edmondson Bob]

Hi, apologies-I forgot to answer the CO question. I find 0.326% CO with 5 deg approach to equilibrium and exit temperature 240 deg C which I think is a reasonable distance from dewpoint at the inlet. I’m guessing since the tube area is unknown and I assume 15 bar(a) steam generation as a safe lower limit for distance from dewpoint at the inlet. Reformer is at 860 deg C outlet, 30 bar(a) and inlet is 3.4:1 S/C with a reasonable lean natural gas feed.

[raj da]

Thanks Bedmonds.

In SMR top fired reformers(such as Linde) where exit of 850C, typically what is the tempereature of flue gas just before it enters the convection section? I am trying to do a heat balance to determine fuel needed. Can it be 850+200 = 1050C? is there any recomended or practical delta temperature for this application?

[Edmondson Bob]

Hi Daraj,
I suspect your 1050 deg fluegas would be quite typical. It’s an interesting design topic. With more reformer tubes area it drops and radiant efficiency rises but as long as the heat can be put to good use (heating reactants, superheating steam, preheating combustion air, driving a feed gas saturator coil, etc a solution ends as a balance of costs. In the example I sent earlier-S/C 3.4 T 860 deg P 30 bars I had a fluegas at 1070 based on the tube area I used. For a case with 2000 kmol/h of lean natural gas we can look at two options:
Tube area 15000 sq ft 25000 sq ft
Fluegas to convection 1151 deg C 1057 deg C
Radiant efficiency 58.4% 63.0%
For these cases I have assumed no pre-reformer but a pre-reformer can be a good option since it soaks up convection bank heat, but if, for example, steam needs to be raised that would be the more optimal solution. It’s hard to zero in on the optimal solution without considering the needs of the balance of the plant. A very interesting subject on which I am not an expert.

[raj da]

Thanks Bedmonds. I think in Linde processes emphasis is on raising steam using the flue gas heat. This HP steam is then used to drive compressors within the process. I believe gas to convection section can be even at 1100 or 1125C.

if i define efficiency of reformer as (heat used for SMR reactions+heat recovered)/(total heat input from fuel burning), do you think this number can be as high as 92 or 93%? thats what a lot of licensors claim these days for a top fired reformer.

[raj da]

i do not have pre-reformer

[raj da]

mine is a methane-rich natural gas(92 mol%) and no use of pre-reformer. purpose is to make hydrogen for ammonia.
Typical conditions are 850-870C reformer exit, S/C of 2.8-3 and 28-32 bara reformer pressure

[Edmondson Bob]

Hi Daraj,
For the conditions you noted using 29 bar, 860 deg, S/C of 2.9 I have 6.45 drygas percent CH4 exit the reformer. With 25000 sq ft of reformer tubes fluegas is 1048 deg for 2000 kmol/h of feed natural gas. This is a big reformer w/ 466 tubes assuming 130 mm OD and 12.2 m heated length. The reformer could be smaller with hotter radiant outlet if use for the heat in the convection bank (mixed feed heating, steam superheating, steam generation, saturator coil etc). I know your hope was CH4 at 5% but to do so required hotter exit or higher (3.4) S/C.

[raj da]

bedmonds, thanks. is 6.45% methane on dry basis? then its not too bad.

you have mentioned some dimensions. i have close to 1350 kmol/hr of natural gas. does that change things? this is close to a 1000 ton per day ammonia plant.

any idea on how much reformer cost for such plants or the duty needed for the SMR?

[astromo]

Regarding the original question of pros/cons:

which is offered by other tech suppliers. I wanted to know what are thre advantages of this design, any thoughts?

… a quick web-search can turn up all sorts of good intel:

[Author]

Posts