Patent Grant
7056399
The present invention relates to a process for treating steels to make them more resistant to coke formation in hydrocarbon processes. Specifically, the method involves a surface treatment process for steels used in transfer line exchangers of steam crackers for ethylene production and in reactors and heat exchangers of refinery processes. Typically, such equipment in contact with hydrocarbon streams are operated at temperatures ranging from 200° C. to 900° C.
In the refinery and petrochemical industry, the most commonly used materials for reactors and heat exchangers are carbon steels due to cost consideration. Often, high alloy steels are used only for hydrocarbon processes where other requirements such as corrosion or operating temperature may become an issue. It is well-known that iron and its oxides present on steel surfaces could act as promoters for coke formation.
Coke formation on equipment surfaces could cause many problems for process operation. Among them, two often mentioned problems are the reduced (distorted) heat transfer across the equipment walls due to the build-up of coke deposits having poor thermal conductivity, and increased pressure drop due to the accumulated coke deposit which can substantially reduce the opening for the process stream and which also increases the surface roughness in contact with hydrocarbon stream. Both of these effects can affect the designed performance of a particular equipment. Other problems with coke formation in hydrocarbon processing equipment include loss of operation time and the required maintenance cost for coke removal using on-line or off-line methods. For example, in transfer line exchangers used for quenching the effluent stream from a steam cracker, coke formation often becomes a major problem restricting furnace run length, especially for naphtha cracking. With emerging technologies for longer furnace run length, coke formation in the transfer line exchangers must be dealt with.
There have been a number of proposals for treatment of steels to reduce their tendency to coke when exposed to hydrocarbons at elevated temperatures. In general, these proposals in the prior art could fall into two categories—the use of coke inhibiting compounds or mixtures to react with the steel surface and form an inert surface prior to its exposure to process hydrocarbons and/or during hydrocarbon processing, and surface passivation through treatment using gases such as hydrogen, carbon dioxides, air or steam prior to exposure to hydrocarbons.
Injection of coke inhibiting compounds or mixtures has become a very popular approach for technology development and to some extent for plant practice.
U.S. Patent Application 20020029514 published Mar. 14, 2002 assigned to Atofina Chemicals Inc. teaches treating a furnace, preferably co-injecting with steam and one or more compounds of the formula R—Sx—R′ where x is an integer from 1 to 5 and R and R′ are selected from the group consisting of a hydrogen atom and a C1-24 straight chain or branched aryl radicals, and one or more compounds of the formula:
wherein R, R′ and R″ are selected from the group consisting of C1-24 straight or branched aryl radicals. The present invention has not only eliminated the hydroxylamines, hydrazines and amine oxides required by the prior art, but also identified additional but essential steps to make the passivation of steel surface more stable.
U.S. Pat. No. 4,636,297 issued Jan. 13, 1987 to Uchiyama et al., assigned to Hakuto Chemical Co., Ltd. teaches applying a mixture of dialkyl thioureas and thiuram mono- and/or di-sulfides in an amount from 10 to 5,000 ppm to the surface of a reactor prone to coke formation. The reference does not teach the specific components used in the present invention nor does it disclose the preliminary reduction nor the curing steps required in the present invention.
U.S. Pat. No. 5,777,188 issued Jul. 7, 1998 to Reed et al., assigned to Phillips Petroleum Company discloses adding to the feed of a steam cracker with steam as a carrier gas and a mixture of polysufides of the formula R—Sx—R′ wherein R and R′ are independent hydrocarbyl radical having 1 to about 30 carbon atoms and x is a number from about 3 to 10. The proposed weight ratio of polysulfides to steam is in the range from about 0.00002:1 to about 1:1. Again the reference fails to teach the specific components used in the present invention nor does it disclose the preliminary reduction and the curing steps required in the present invention.
In addition, there are many other chemicals or mixtures of them that could be used for reduction of coke formation under cracking and TLE operating conditions. Tong et al. has claimed a number of organic phosphorous compounds (U.S. Pat. Nos. 5,354,450; 5,779,881; 5,360,531 and 5,954,943, assigned to Nalco/Exxon) that can be used as coke inhibitors for coke reduction under coil and TLE conditions. A combination of gallium, tin, silicon, antimony, and aluminum has also been claimed in the prior art (U.S. Pat. Nos. 4,687,567; U.S. 4,692,234; and U.S. 4,804,487), assigned to Phillips Petroleum. Additionally, certain inorganic salts, a mixture of Group IA and IIA metal salts and a boron acid (U.S. Pat. No. 5,358,626) assigned to Tetra International, have been claimed as effective in coke reduction under coil conditions. Once again, these references fail to teach the specific components used in the present invention nor do they disclose the preliminary reduction nor the curing steps required in the present invention.
The other group of methods or processes available in the prior art, teaches the use of gases, such as H2, carbon oxides, steam and air to treat steel surfaces prior to their exposure to hydrocarbon process streams in order to minimize the coking propensity of steel surfaces.
U.S. Pat. No. 5,501,878 issued Mar. 26, 1996, assigned to Mannesmann Aktiengesellschaft; KTI Group B.V. teaches treating the surface of a heat exchanger which comes in contact with hydrocarbons with a mixture of steam and 5 to 20 weight % hydrogen at a temperature from about 400° C. to 550° C. for from 1 to 6 hours to reduce Fe2O3, that is catalytically active to produce coke, to Fe3O4 that is not as active to produce coke. The present invention uses a lower amount of hydrogen than that specified in the reference and comprises further steps not disclosed in the reference.
U.S. Pat. No. 6,436,202 issued Aug. 20, 2002, assigned to NOVA Chemicals teaches a process for treating stainless steel comprising from 13–50 weight % Cr, 20–50 weight % Ni and at least 0.2 weight % Mn in the presence of a low oxidizing atmosphere, which comprises from 0.5 to 1.5 weight % of steam, from 10 to 99.5 weight % of one or more gases selected from the group consisting of hydrogen, CO and CO2 and from 0 to 88 weight % of an inert gas selected from the group consisting nitrogen, argon and helium. In an earlier U.S. Pat. No. 5,630,887, again assigned to NOVA Chemicals (previously NOVACOR Chemicals) a similar procedure was proposed for the treatment of stainless steel furnace tubes which are used in the petrochemical industry. This treatment involves exposing stainless steel to an atmosphere containing a low amount of oxygen at temperatures up to 1200° C. for up to about 50 hours. The stainless steel treated according to such a procedure will have a lower tendency to coke formation during use. However, these treatments are not suggested for steels with a Cr content less than 13 weight %, for instance, carbon steel, which comprises typically less than 5 weight % Cr. In addition, the required use of the coke inhibiting compounds of the present invention and the curing step have not been disclosed in these references.
The present invention seeks to provide an effective method of treating a steel, preferably but not limited to carbon steels, subject to conditions where coke is likely to form to reduce coke formation.
The present invention provides a process for treating a steel comprising not less than 35 weight % Fe, comprising:
(i) reducing the surface of the steel by contacting it with a mixture comprising from 0.001 to 4.9 weight % of H2 and 99.999 to 95.1 weight % of one or more gases selected from the group consisting of inert gases (such as argon, nitrogen, helium, etc.) and steam at a temperature of from 200° C. to 900° C. and a pressure from 0.1 to 500 psig for a time from 10 minutes to 10 hours;
(ii) treating the reduced surface of the steel with a composition comprising:
the sum of components (a) through (f) being adjusted to a total 100 weight %,
in an amount from 10 to 10,000 ppm in a carrier gas selected from the group consisting of steam, inert gases and hydrocarbons at a temperature from 400° C. to 850° C. for a time from 10 minutes to 10 hours; and
(iii) curing the resulted surface in a carrier gas selected from the group consisting of steam, and inert gases (such as argon, nitrogen and helium) or a mixture thereof for a time from 0.1 to 50 hours.
The present invention relates to the treatment of steels, particularly but not limited to carbon steels, including steels with a Fe composition of at least 35 weight % (wt %) (i.e. from 35 to 100 wt % Fe), preferably 60 to 100 wt %, most preferably 80 to 100 wt % Fe. This will include HK, HP steel alloys, but not higher grade steel alloys. The classification and composition of such steels are known to those skilled in the art.
One type of stainless steels which may be used in accordance with the present invention broadly comprises: from 10 to 45, preferably from 12 to 35 weight % of chromium and at least 0.2 weight %, up to 3 weight % preferably not more than 2 weight % of Mn; from 20 to 50, preferably from 25 to 48, weight % of Ni; from 0.3 to 2, preferably 0.5 to 1.5 weight % of Si; less than 5, typically less than 3 weight % of titanium, niobium and all other trace metals; and carbon in an amount of less than 0.75 weight %. The balance of the stainless steel is substantially iron.
A complete treatment procedure consists of a preliminary reduction step of the steel surface, a passivation step involving the use of coke inhibiting compounds and their mixtures, and a curing period using steam and one or more of inert gases to stabilize the already passive steel surfaces. This treatment procedure may be carried out on the steel in situ (e.g. in a cracker or a reactor for a hydrocarbon process) as well as externally such as an off-site treatment.
In the first step of the present invention the steel is reduced typically using H2 mixed with one or more gases selected from the group consisting of inert gases such as argon, nitrogen, helium etc., and steam and mixtures thereof. Preferably the gas is steam. Generally, the steel surface is treated with hydrogen in steam alone or optionally together with some of the inert carrier gas such as argon, nitrogen, helium etc. The hydrogen may be present in the carrier gas in an amount from 0.001 to 4.9, preferably 0.01 to 2, most preferably 0.1 to 1 weight %.
The treatment is carried out at temperatures from 200° C. to 900° C. preferably 300° C. to 800° C., most preferably from 300° C. to 700° C.; and at pressures from 0.1 (0.689 kPa gage) to 500 psig (3.447×103 kPa gage), preferably from 0.1 to 300 psig (2.068×103 kPa gage), most preferably from 0.1 to 100 psig (6.89×102 kPa gage) for a time from 10 minutes to 10 hours, preferably from 30 minutes to 5 hours, most preferably from 1 to 3 hours.
During the second step of the present treatment procedure, several coke inhibiting compounds and mixtures thereof may be used to passivate the steel surface so that the treated steel has less of a tendency for coke formation. The composition of the coke inhibiting compounds used comprises:
These coke inhibiting compounds or mixture may be carried onto steel surface by a carrier medium selected from the group consisting of inert gases such as argon or nitrogen, or steam, or light hydrocarbons such as methane or ethane, or a mixture thereof, in an amount from 10 to 10,000 ppm (weight), at a temperature from 300° C. to 850° C. for a time from 10 minutes to 10 hours, preferably in an amount from 20 to 5,000 ppm (by weight), most preferably in an amount from 30 to 2,000 ppm (by weight (e.g. wppm) preferably at a temperature from 300 to 800° C. for 30 minutes to 5 hours.
In accordance with the present invention, the resulting steel surface should be further treated by following a curing procedure, which may consist of passing steam alone or steam mixed with one or more inert gases such as argon or nitrogen at a steam concentration no less than 2 wt %. This curing process may be carried out at a temperature between 200° C. and 900° C., preferably 300° C. to 800° C. for a period of 0.1 to 50 hours, preferably 0.5 to 20 hours at steam partial pressures from 0.1 (0.689 kPa gage) to 100 psig (68.95 kPa gage), preferably from 0.1 to 60 psig (413.7 kPa gage), most preferably from 0.1 to 30 psig (206.8 kPa gage).
The steels treated in accordance with the present invention may be used in processing a number of types of hydrocarbons including lower C1-8 alkanes such as ethane, propane, butane, naphtha, vacuum gas oil, atmospheric gas oil, and crude oil. Preferably, the hydrocarbons will comprise a significant amount (e.g. greater than 60 wt %) of C1-8 alkanes, most preferably selected from the group consisting of ethane, propane, butane and naphtha.
The steel treated in accordance with the present invention may be used in a number of applications where a hydrocarbon will be exposed to the steel at relatively mild temperatures typically at temperatures from 300° C. to 800° C. One use for the steels treated in accordance with the present invention is in the transfer line exchanger (TLE) at the outlet of a coil of a steam cracking furnace.
The present invention will now be illustrated by the following non-limiting examples. In the examples either or both of a thermogravimetric testing unit (TGTU) used in the examples and a tubular cracking and quenching reactor (TCQR) may be used.
The thermogravimetric testing unit (TGTU) is illustrated in
The schematic of TCQR is shown in
A series of sample powders of Fe containing compounds (listed in Table 1) were tested under simulated ethane cracking conditions at 840° C. in the TGTU. Initially, the TGTU furnace was heated at a rate of 15° C./min in a flow of N2 purge at 25 sccm (standard cubic centimeters per second). When the temperature reached 840° C., ethane was admitted via the wet route at 15 sccm and cracked in the cracking zone (7 of
The results show that sulfides have the lowest coking rates while the oxides show substantially higher coking rates under the same testing condition. The maximum coke formations of these compounds occur typically at the beginning of ethane cracking.
A series of H2 reduction tests were carried out using the TGTU. The same powder samples, placed in the homogeneous temperature zone (7 in
Fe
Fe
Fe
Fe
Fe
Fe
These results show that Fe oxides can be more easily reduced using wet H2 than the sulfides, with generally lower upper temperatures for the oxides than for the sulfides. For the two sulfides tested, the reduction reactions did not appear to have reached completion at a temperature up to 900° C. and with 30 minutes hold time. Additionally, Fe3O4 was observed to also reach close to 900° C. for a complete reduction. Such a difference could be attributed to possible differences in crystalline structure between the sample Fe3O4 and the intermediate product Fe3O4 converted from Fe2O3.
For comparison, three experiments were carried out in the TGTU using carbon steel coupons (A387F22) of 0.187″×0.48″×0.96″ in size. The coupons with fresh surfaces polished to 600 grit were placed at the 600° C. position in the TGTU furnace which was maintained at 840° C. with a feed gas flowing through the wet route during the experiments. In one of the experiments, one of the coupons was heated in wet N2 to 600° C. (840° C. furnace temperature) and air flowing at 50 sccm was introduced into the furnace to oxidize the coupon surface for 1 hour, which was to simulate a wet decoke in ethylene plant. Afterwards, dimethyl disulfide vapour was carried in by purging N2 at 50 sccm through the wet route for surface sulfiding of the coupon. Then ethane was introduced into the furnace for steam cracking for 1 hour to determine the coking rate. With the other coupon, an H2 reduction step took place after the oxidation for 1 hour and a steam curing step took place after sulfiding for another hour. The results from both experiments are given in Table 3.
(*)step not executed in the run.
(**)S concentration in the gas feed to TGTU furnace is about 0.45 wt %.
The results show that significant reduction (68%) in coking rate can be achieved by sulfiding alone at a high S concentration. However, adding both H2 reduction prior to sulfiding and steam curing after sulfiding can reduce coke formation further up to 95%.
Ethane steam cracking tests were carried out in the TCQR with A387F11 carbon steel coupons placed in the TLE section, at positions described previously. Ethane was steam cracked in the furnace at 900° C. (wall temperature) with the residence time at about 1 second. The steam to hydrocarbon ratio was maintained at 0.3 (w/w) and the tests lasted for 10 hours. Based on product analyses from a gas chromatograph, ethane conversion was about 65 wt %, throughout the 10 hours experimentation period. A coke inhibitor consisting of 10 wt % DMDS, 70 wt % TBPS, 10 wt % PTMP and 10 wt % DMP was injected at the simulated TLE inlet at various concentration. The results are listed in Table 4. As a comparison, results from two baseline runs are also included.
The results in Table 5 show that by using the passivation procedure (H2 reduction, surface modifier injection and steam curing), the reduction in total coke formed in the simulated TLE section are in the range up to 76.9 wt %. Inhibitors injected at higher concentration are observed to cause more coke formation at lower temperature (such as at 550° C.) section and therefore, the total coke reduction is affected. Otherwise, inhibitors injected at a concentration between 300 to 650 wppm for about 1 hour are found to give the best results in coke reduction.
Three experiments were carried out in the TCQR using a naphtha feed collected from a NOVA Chemicals' plant at Corunna. This naphtha was fed into TCQR at 0.19 kg/hr with steam feeding at 50 wt % of the naphtha feed. The cracking furnace was maintained at 850° C. with a residence time at about 1 second. Under such a condition, the conversion of naphtha was about 65 wt % based on gas chromatograph analyses. The overall reaction time for each experiment was maintained for 6 hours. For each experiment, four fresh carbon steel coupons (A387F22) were placed in the simulated TLE section at positions as described previously. Once the cracking furnace reached 850° C. under N2 purge, a steam ramping step was carried out to warm up the TLE section to its desired temperature profile. Thereafter, an oxidation step took place with the purging N2 replaced by air for 60 minutes. This step was to create an oxide layer on the coupon surfaces, simulating plant decoke operation. Afterwards, the coupons went through the passivation steps of reduction, inhibitor injection and steam curing as shown in Table 5. For comparison, a baseline run was carried out without these three steps.
The results (Table 5) show that the overall reduction in coke are 29.9 wt % and 17.2 wt % for test-1 and test-2, respectively, which are much less than the coke reduction observed from ethane cracking experiments (Example 4). However, it is also noted that the reductions in coke formation at higher temperatures are much higher than those at lower temperatures. For instance, at 650° C., the coke reduction is about 75 wt %, while the numbers for 550° C. and 450° C. are 69.7 wt % and 54.5 wt %, respectively. At 350° C., there is very little reduction, if any, in coke formation. This phenomenon is likely a reflection of the difference between coke formed at higher temperatures and at lower temperatures. Often condensation coke is believed to form at low temperatures, such as 350° C., and the formation rate of such coke (or tar) is not sensitive to surface properties. However, at higher temperatures, coke is believed to form through catalytic mechanisms and therefore the formation rate is sensitive to surface properties, such as the presence of coke promoting oxides.
(*)inhibitor used for this test contained 5 wt % DSFM, 5 wt % DMP, 20 wt % DMDS, 50 wt % TBPS and 10 wt % PTMP.
| Number | Name | Date | Kind |
|---|---|---|---|
| 4636297 | Uchiyama et al. | Jan 1987 | A |
| 4687567 | Porter et al. | Aug 1987 | A |
| 4692234 | Porter et al. | Sep 1987 | A |
| 4804487 | Reed et al. | Feb 1989 | A |
| 5354450 | Tong et al. | Oct 1994 | A |
| 5358626 | Gandman et al. | Oct 1994 | A |
| 5360531 | Tong et al. | Nov 1994 | A |
| 5501878 | Barendregt et al. | Mar 1996 | A |
| 5630887 | Benum et al. | May 1997 | A |
| 5777188 | Reed et al. | Jul 1998 | A |
| 5779881 | Tong et al. | Jul 1998 | A |
| 5954943 | Tong et al. | Sep 1999 | A |
| 6436202 | Benum et al. | Aug 2002 | B1 |
| 20020029514 | Lindstrom | Mar 2002 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20040216815 A1 | Nov 2004 | US |
