Mineral oils: processes and products – Chemical conversion of hydrocarbons – With prevention or removal of deleterious carbon...
Reexamination Certificate
1999-03-12
2001-02-06
Yildirim, Bekir L. (Department: 1764)
Mineral oils: processes and products
Chemical conversion of hydrocarbons
With prevention or removal of deleterious carbon...
C208S425000, C208S130000, C585S648000
Reexamination Certificate
active
06183626
ABSTRACT:
SUMMARY OF THE INVENTION
The invention relates to a process for flexible steam cracking of hydrocarbons, i.e., a process that is compatible with a wide variety of feedstocks to be cracked and a wide variety of operating conditions. It also relates to a process for decoking of the steam-cracking unit.
BACKGROUND OF THE INVENTION
The technological background is illustrated by patent applications WO-A-90.12851, WO-A-96.20255, EP-A-0 036 151, EP-A-0 036 609, EP-A-0 272 378, and FR-A-2 647 804.
The steam-cracking process is the basic process of the petrochemical industry and consists in cracking a feedstock of hydrocarbons and water vapor at high temperature and then abruptly cooling it. The main operating problem arises from the deposition of carbon-containing products on the inner walls of the unit. These deposits, which consist of coke or heavy pyrolysis tars that are condensed and more or less agglomerated, limit heat transfer in the cracking zone (in a pyrolysis pipe coil) and the indirect quenching zone (effluent quenching exchanger), thus requiring frequent shutdowns to decoke the unit.
The standard cycle periods (operation between two complete chemical decokings of the cracking zone, with air and/or with vapor) are either set (scheduled shutdowns) or variable, depending on the coking of the unit, and generally are spread between 3 weeks and 12 weeks for feedstocks such as naphtha and liquefied petroleum gases.
It is known to one skilled in the art that the coking problems that are encountered during the cracking of heavy feedstocks (atmospheric gas oils, heavy gas oils, distillates under vacuum) are much more serious than those that are encountered on standard feedstocks, such as naphtha.
Consequently, these feedstocks cannot be cracked in standard steam-crackers that are designed for cracking naphtha, and cannot be cracked according to the known processes that in special furnaces typically comprise direct quenching (with pyrolysis oil) of the steam-cracking effluents; this considerably impairs the energy balance of the unit (no high-pressure vapor production).
The known processes that make it possible to have flexibility with respect to the heavy feedstocks are therefore incompatible with the existing steam-cracking units on standard feedstocks, and they have a greatly degraded energy balance.
Furthermore, in the case of relatively light feedstocks, there is a tendency also to use as steam-cracking feedstocks poor-quality naphthas such as C
4
, C
5
recycling fractions, thermal cracking gasolines, or the core of fraction of catalytic cracking (FCC). These olefinic feedstocks lead to significant coking problems, in particular with a high degree of severity.
Furthermore, the applicants have already proposed (EP-A-419 643, EP-A-425 633 and EP-A 447 527) a process for in-service decoking of steam-cracking units, by injecting erosive solid particles in order to solve coking problems and to obtain continuous or approximately continuous steam cracking (for example, cycle periods on the order of 1 year).
For a given feedstock, this process consists in allowing a layer of coke to form and mature on the inner walls of the cracking coil and then injecting erosive particles (for example, mineral particles that are hard, with a diameter of less than 150 micrometers, spherical or angular) in sufficient quantity to stabilize approximately the coking state of the pipes without totally eliminating the pre-layer of coke, which plays a protective role for these pipes.
This process requires good knowledge of the coking speeds of the feedstock that is being considered and a coil design such that there is a certain correspondence between the local coking speeds, which are related to the progress of cracking along the coil, and the intensity of erosion, which is related to the speed profile along the coil and to the nature of the erosive particles. With, on the one hand, simulations of coking speeds and the profile of circulation speeds in the coil and, on the other, pilot experiments, it is possible to provide for approximately continuous steam-cracking conditions of the feedstock under study.
It can be ensured that there is very little if any erosion of the pipes, and erosion can be monitored by the analysis of metal traces (iron, chromium, nickel) in the powders that are recovered.
The applicants have therefore sought to improve this process, which can be applied to the cracking of a given feedstock, in the case of a flexible furnace that can successively process a large number of different feedstocks, with variable operating conditions (flow rate, degree of dilution, degree of severity of cracking).
Pilot tests have been carried out and have provided several unexpected results:
It was actually found that the initial coking of the coil (at the beginning of the cycle) could vary to a very significant extent depending on the feedstock, including for feedstocks that are slightly different in terms of chemical composition but are of different origins. It has not been possible to explain this completely, and it may be due to impurities that are present in the feedstock.
Furthermore, the effectiveness of decoking has proven to depend in particular on the feedstocks and operating conditions (different nature of the coke). In particular, it was found that at the beginning of the reaction zone the light feedstocks: C
3
, C
4
, light naphtha, produce a much more fragile catalytic coke (by 5 to 10 times) than the asymptotic coke that predominates in the middle and at the end of the reaction zone. For these feedstocks it is therefore desirable to limit the circulation speed in this zone to maintain a protective coke layer and/or to avoid risks of erosion of the cracking pipes.
Thus, it has not been possible to predetermine the quantities of particles that are suitable for each feedstock and each operating condition without preliminary tests, and such tests cannot be carried out in the case of a flexible industrial furnace. In addition, with regard to the prevention of erosion risks, the geometry of the cracking reactor that is suitable for a given feedstock is not the same as that which is suitable for another feedstock that has a degree of dilution and a type of coke that are different, for which the appropriate profile of circulation speeds will be different.
Finally, because of the difficulties in obtaining reliable and specific measurements of skin temperatures of the pipes by optical pyrometry, as well as due to fluctuations of these temperatures and the loss of load under the variable operating conditions, it is very difficult to monitor effectively the state of coking of the pipe without resorting frequently to a constant reference state, which cannot be done for a flexible industrial furnace, and thus to be able to control in real time the coking of a pyrolysis coil.
The process has therefore proven difficult to use industrially under variable operating conditions, and it has not been possible to keep any trace of erosion from arising in cracking pipes for all the pilot tests.
It thus appeared that the continuous steam-cracking process could not be adapted in the case of a flexible furnace and should be reserved solely for cracking of identical or similar feedstocks under relatively stable conditions.
Furthermore, the applicants have noted that elimination of deposits in the indirect quenching exchanger could be achieved much more easily than in the pyrolysis pipes and that, even in the case of injection of particles in excess quantity, no erosion was noted.
Thus, it appeared, surprisingly enough, that the carbon-containing deposits of the quenching exchanger, in particular in the case of heavy feedstocks, were much more fragile than the coke of the cracking pipes. It was found, actually, that the brittleness, compared to the erosion by the solid particles under test, was at least 25 times greater for the coke of the quenching exchanger than for the asymptotic coke of the pyrolysis pipes.
The absence of erosion that is noted for the pipes of the exchanger themselves is explained by the fact that
Burzynski Jean-Pierre
Courteheuse Gerard
Gougne Yves
Huin Roland
Lenglet Eric
Institute Francais du Petrole
Millen White Zelano & Branigan P.C.
Yildirim Bekir L.
LandOfFree
Method and device for steam cracking comprising the... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and device for steam cracking comprising the..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and device for steam cracking comprising the... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2575670