[astromo]

astromo wrote:

Refer here for one supplier’s explanatory detail:

Check the video and note the use of an on-site furnace to heat the material and expand it to “switch on” perlite’s insulating properties. Think of corn being heated to become popcorn and its low density, high void:surface area.

For some reason that I can’t fathom, the forum software is not displaying the URL link that I posted (at least for me anyway), even though it shows up when quote the post.

Just in case I’m not alone, the supplier and product I referred to is Imerys Cryolite 200. FWIW, I’ve got no affiliation with that company and have no idea whether their product is good or bad. However, they have posted a handy video that explains the process for onsite preparation for cold insulation use. That’s why I used it in the reply.

[astromo]

astromo wrote:

From what I know, this process was patented by Bernard Grotz, Jr., back in 1969 for the CF Braun company:

Refer Figure 2 and the supporting text in the patent.

CF Braun made its way through various hands and is now part of KBR, who have adopted the “purifier” process as a core part of their hydrocarbon fed ammonia technology.

If you can find this paper:
Gosnell, J. & Malhotra, A., KBR’s New PURIFIERplus Process, Nitrogen Conference, Vienna, Austria, 12-15 March 2006
It will give you an update of development of the process in the intervening 35 years but the core tech was pretty well in the same basic state after it left the hands of Grotz. So, that’s a clearly positive testament to the robustness of the original design.

I’ll leave you to review the detail. The answer to your question is covered by the patent.

For the Grotz patent, search for:
US3442613 Grotz, CF Braun, Hydrocarbon Reforming for Production of a Synthesis Gas from which Ammonia can be Prepared

[astromo]

Regardless of the insulating duty, perlite needs to stay dry to be effective. So, single wall tanks and cold boxes are purged with high purity N2 to maintain dryness. As long as the purge system is maintained, my understanding is that perlite will maintain its insulating performance for years cum decades of service. I’d expect some top up could be needed to account for some settlement over time but an experienced perlite supplier should be able to answer that question in depth. Plus, I’d expect that the smart move is to load a bit of excess to account for settlement.

Refer here for one supplier’s explanatory detail:

Check the video and note the use of an on-site furnace to heat the material and expand it to “switch on” perlite’s insulating properties. Think of corn being heated to become popcorn and its low density, high void:surface area.

If you can source it, this paper:
McGrath, P. & Tapp, P., Ice Removal from the Annular Space of an On-line Atmospheric Ammonia Storage Tank, 54th Annual Safety in Ammonia Plants and Related Facilities Symposium, 2009
reviews the experience of dealing with wet perlite (that then sets solid) in the annular space of a single wall tank.

If you’re thinking of external foam insulation then one key disadvantage that polyurethane foam (PUF) class of insulation carries is its flammability. Perlite is inert in that regard, so that’s a positive option. Another possibility is to use cellular glass, which from what I know is a better performer in terms of moisture resistance and without doubt, will not go up in flames like PUF. Downside of cellular glass is that it’s comparatively more expensive.

[astromo]

From what I know, this process was patented by Bernard Grotz, Jr., back in 1969 for the CF Braun company:

Refer Figure 2 and the supporting text in the patent.

CF Braun made its way through various hands and is now part of KBR, who have adopted the “purifier” process as a core part of their hydrocarbon fed ammonia technology.

If you can find this paper:
Gosnell, J. & Malhotra, A., KBR’s New PURIFIERplus Process, Nitrogen Conference, Vienna, Austria, 12-15 March 2006
It will give you an update of development of the process in the intervening 35 years but the core tech was pretty well in the same basic state after it left the hands of Grotz. So, that’s a clearly positive testament to the robustness of the original design.

I’ll leave you to review the detail. The answer to your question is covered by the patent.

[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:

[astromo]

To get started I’d look to the advice from Max Appl’s 1999 book, “Ammonia Principles and Industrial Practice“:

4.2.1.3. Intermediate-Temperature Shift (ITS)
A relatively new process concept is the intermediate temperature shift [627], [628], which performs the reaction in a single step. The catalyst is based on a copper-zinc-alumina formulation and optimized for operating in a wider temperature range (200 – 350 “C) than the standard LTS catalyst (190 – 275 “C). The reaction heat can be removed by use of a tube-cooled reactor raising steam or heating water for gas saturation to supply process steam in the reforming section (Linde LAC, ICI Catalco LCA). In a new plant using the spiral-wound Linde reactor [629], a methane slip of only 0.7 mol % (dry basis) is achieved.

However, Appl is in error regarding the reference to “methane slip”. Referring to the source document [629], Ilg & Kandziora, Linde Ammonia Concept, 41st AIChE Ammonia Safety Symposium, Boston 1996; Ammonia Plant Safety vol.37 (1997) pp.341-352, on p. 344, the following is noted:

Linde Isothermal Reactor

This reactor (Figure 8) is used for the CO shift reaction, and allows conversion to below 0.7% CO (dry basis) in a single step.

The reaction involved is primarily water gas shift, however, as with any shift reaction there are potential side reactions, such as methanol formation and Fischer-Tropsch. One of the claimed benefits of an iso-shift reactor is that the avoidance of HT shift conditions reduces the quantity of contaminants formed by side reaction.

[astromo]

Useful cross reference for info:
Hygrogen Recovery System – Prism Membrane

• This reply was modified 1 year, 10 months ago by Petrus.

[astromo]

I know that this one is old but just to round things out, all that I could suggest is from ISO 15859-11:2004, Space Systems — Fluid Characteristics, Sampling and Test Methods — Part 11: Ammonia. It doesn’t call out Fe as a standalone impurity but has a lumped parameter, Nonvolatile Residue Content. That standard specifies that the analysis identify the residue constituents by the infrared spectrometric method.

I can see by web search that iron supplement tablet are often analysed using this technique. The iron is readily dissolved in acid and then the solution is subject to IR spectrometry. I’d expect that calibration of the spectro is done using standard solutions of known Fe content.

From what I can tell, an analytic method to determine Fe content should be relatively simple to develop and achieve accurate and repeatable results.

[astromo]

Twigg’s book, Catalyst Handbook, published by CRC Press (2nd Ed., ca. 1999, is the latest to the best of my knowledge) is my go to reference for these kind of questions.

Twigg confirms the application of Le Chatelier’s principle in the chapter on water-gas shift reaction, noting that the “position of equilibrium is virtually unaffected by pressure“. Twigg expands further with:

quote :

Under adiabatic conditions conversion in a single bed of catalyst is thermodynamically limited – as the reaction proceeds the heat of reaction increases the operating temperature, and so restricts the conversion possible.

So, after the system is bounded or constrained in terms of pressure then temperature, for a given catalyst’s activity and a fixed volume, for a given inlet concentration of CO, what have you got left to drive the reaction forward? I can’t help but be drawn to the left hand side of the reaction for an answer.

But as with most things in life, Goldilocks had the right idea in terms of not too much and not too little.

[astromo]

To concur with the good advice cautioning that fluid temperature and auto-ignition is not the only mechanism by which rich H2 can ignite, I’d suggest that those interested consult the advice found in API STD 521, Pressure-Relieving and Depressuring Systems. My 7th Ed. copy at 5.8.4.2.3 Release of Hydrogen-rich Streams recounts empirical work by NASA and notes that energy as low as 0.017 mJ has been found to ignite H2. Iron/Iron Oxide particles, such as that found in a vent, can also provide a source of ignition.

In short, in practice hydrogen is extremely easy to light off.

[astromo]

Regarding the OP’s questions:

Reply to Q1. Yes you can run your plant but you will be cracking the organic sulphur compounds across the primary reformer and progressively poisoning it – so not a good plan (no surprise with that reply).

Reply to Q2. I would ask your catalyst supplier. Depending on the load spec, you will get some variation in the answer for how long you can run. It’s not unusual for the initial section of the tubes to be loaded with a sulphur resistant catalyst, as a precaution. So, it’s not just a question of make of catalyst but specification and loading as well.

I ran a plant (see comment below) where H2 cylinders were designed in for start up to run the hydro-desulphuriser. Too often, we ran out. The gas supplied to the site was close to the gas field and far upstream of domestic users. Because of this the sulphur level in the pipeline gas was virtually zero, however due to poor design planning (gas detection could have avoided the installed solution), an odourant injection station was specified at the custody transfer point. Then, about 200 m of pipe later, the mercaptan odourant was removed by the HDS. With the low level of odourant injected, we routinely started up without backup H2 supply but we checked this plan with the supplier. Procedurally, the odourant was switched off during this period. This was largely of psychological benefit because over time the mercaptan will infuse into the pipework and then waft out but it was better than leaving it continue to be squirted in to the feed. While I was there, we did not observe any deleterious effects but an issue to watch as part of plant performance management.

Some further thoughts for being able to treat the issue:
1. For the scenario in question, a stand alone catalyst unit could be used to crack NH3 to H2 + N2 to avoid relying on a recycle H2 compressor. This kind of tech is gaining global interest due to “green H2” and “green NH3” and variants.
2. Another option could be an activated carbon bed for start up until the H2 compressor is returned to service. In normal operation, bypass the unit. Take care to consider the cost of media disposal.
3. A 3rd option I’ve seen is the use of H2 cylinders. For the case in question, the system was significantly undersized and relied on manual control. Poorly conceived with regard to the plant I’m thinking. The key downside of this concept is the need to import H2 rather than have a manufacturing method immediately to hand.

For 2 and 3, how long you can run is a system endurance / sizing basis trade-off. Break out your calculator. Plus there’s a net present cost calculation to work out which option is likely to be the best value.
For 1, most ammonia plants would have sufficient storage of liquid NH3 to supply the HDS needs for the period of the H2 compressor’s mean time to repair. Once the capital cost is addressed, it’s then just a question of the operating cost.

[astromo]

In my earlier days, I was part of a team to commission a vibrating prill head for low density AN production.

The claimed benefit of this technology over a static unit was supposed to be a reduction in the formation of fines and “micro prills”. The oscillation was supposed to help form a clean break of the melt stream from the prill head “bucket” holes.

So, as with many decisions in the engineering world, the balance to consider is the compromise between minimal moving parts, less to go wrong over improved production performance. Apologies, I don’t have the comparative, quantitative data to point to. I leave it to the OP to research further detail.

It would be useful to see where that investigation concludes.

[astromo]

quote vpattabathula post_id=1423 time=1601332230 user_id=2609:

In modern day NH3 plant design, all process vents should be connected to flare headers. Its considered as a best practice to have separate flares for process gas (typically front-end) and NH3 (back-end).

To pick up on this point, the importance of backpressure on relief design (especially conventional PSVs) is another reason for separating front end relief systems from the back end given the system design pressure difference between the two parts of the plant.

[astromo]

As a post-script, if the ammonia is used as feedstock to a nitric acid plant, the potential to poison the catalyst gauze needs to be considered. The maximum limit that I’ve come across before is 5 ppm (wt), in line with something like this:
https://www.hillbrothers.com/pdf/Product-Profiles/Commercial-Grade.pdf

Many refrigeration circuits use oil flooded screw compressors, so at first pass, a specification of 2 ppm reads to be rather stringent to me.

Tighter specifications can be found, such as:
https://www.linde-gas.com/en/images/linde-datasheet-01-ammonia-June-2017_tcm17-417364.pdf
but that’s for use in a sensitive industrial process, such as the production of Si3N4 bearings and other high performance ceramics.

Fertilizer grade (i.e. direct application) anhydrous ammonia is specified to a nominal 20 ppm:
https://www.nagarjunafertilizers.com/solutions_snb_products.htm
or just doesn’t rate a mention, such as:
https://iowafertilizer.com/assets/uploads/2015/09/Anhydrous-Ammonia-Fertilizer-Grade-Spec-Sheet-1.pdf

So, the oil quality required will depend on the end use. So, this is a question to ask your customer.

[astromo]

A number of ammonia plants dating back to the 60s and 70s, that are still operating, were designed with vents rather than flares for PSV relief disposal.

Carbamate formation due to NH3+CO2 reaction can block vent headers and compromise relief discharge rate. I would place this factor as the key concern for avoiding combination of the two species above corrosion potential.

Vents were commonly used decades ago instead of flares because (amongst other factors) they are – cheap to build and operate, greenhouse gas emissions were of little concern and the potential for toxic gas release was less concerning to the local community.

Arguably, flares represent “best” practice and (in my opinion) should be considered as the baseline approach for hazardous gas disposal for new designs.

[Author]

Posts