Thermal cracking vaporization unit construction
Kind Code:

In the process of employing a liquid whole crude oil and/or natural gas condensate in a thermal cracking process that uses a vaporization/mild cracking unit upstream of a thermal cracking furnace, a martensitic steel is employed in the construction of the unit and the unit is subjected to a de-hydrogenation treatment.

Kirkham, Kenneth K. (Humble, TX, US)
Cleavinger, James G. (Baytown, TX, US)
King, Ralph E. (Friendswood, TX, US)
Application Number:
Publication Date:
Filing Date:
Primary Class:
International Classes:
View Patent Images:
Related US Applications:

Primary Examiner:
Attorney, Agent or Firm:
LyondellBasell Industries (Houston, TX, US)
We claim:

1. In a method for the processing of at least one of a whole crude oil and a natural gas condensate by a combination of vaporization and mild thermal cracking in a pressure vessel upstream of a thermal cracking furnace, the improvement comprising constructing said pressure vessel from at least one iron based martensitic steel containing minor amounts of chromium, molybdenum, and vanadium.

2. The method of claim 1 wherein said steel nominally contains about 9 wt. % chromium, about 1 wt. % molybdenum, and about 0.02 wt. % vanadium.

3. The method of claim 1 wherein said steel is grade 91—ASME Section 2A SA387.

4. In the construction of a vaporization/mild thermal cracking pressure vessel, the improvement comprising constructing said vessel from an iron based martensitic steel containing minor amounts of chromium, molybdenum, and vanadium, employing welding to construct said vessel, and subjecting said vessel to post-welding heat treating.

5. The method of claim 4 wherein said steel nominally contains about 9 wt. % chromium, about 1 wt. % molybdenum, and about 0.02 wt. % vanadium.

6. The method of claim 4 wherein said vessel in its entirety as constructed is subjected to said post-welding heat treatment.

7. The method of claim 6 wherein said de-hydrogenation treatment comprises heating said vessel at a temperature of at least about 600 F for at least about 2 hours.



1. Field of Invention

This invention relates to the thermal cracking of whole crude oil, natural gas condensate and/or mixtures thereof using a vaporization/mild cracking unit upstream of a thermal cracking furnace. More particularly, this invention relates to the construction of a vaporization/mild cracking unit from a martensitic steel not heretofore used in pressure vessel construction.

2. Description of the Prior Art

Thermal cracking (pyrolysis) of hydrocarbons is a non-catalytic petrochemical process that is widely used to produce olefins such as ethylene, propylene, butenes, butadiene, and aromatics such as benzene, toluene, and xylenes.

Basically, a hydrocarbon feedstock such as naphtha, gas oil or other fractions of whole crude oil that are produced by distilling or otherwise fractionating whole crude oil, is mixed with steam which serves as a diluent to keep the hydrocarbon molecules separated. The steam/hydrocarbon mixture is preheated to from about 900 to about 1,000 degrees Fahrenheit (° F. or F), and then enters the reaction zone where it is very quickly heated to a severe hydrocarbon thermal cracking temperature in the range of from about 1,450 to about 1,550 F. Thermal cracking is accomplished without the aid of any catalyst.

This process is carried out in a pyrolysis furnace (steam cracker) at pressures in the reaction zone ranging from about 10 to about 30 psig. Pyrolysis furnaces have internally thereof a convection section and a radiant section. Preheating is accomplished in the convection section, while severe cracking occurs in the radiant section.

After severe thermal cracking, the effluent from the pyrolysis furnace contains gaseous hydrocarbons of great variety, e.g., from one to thirty-five carbon atoms per molecule. These gaseous hydrocarbons can be saturated, monounsaturated, and polyunsaturated, and can be aliphatic, alicyclics, and/or aromatic. The cracked gas also contains significant amounts of molecular hydrogen (hydrogen).

Thus, conventional steam (thermal) cracking, as carried out in a commercial olefin production plant, employs a fraction, not the entirety, of whole crude oil, and totally vaporizes that fraction while thermally cracking same in a cracking furnace. The cracked product can contain, for example, about 1 weight percent (wt. %) hydrogen, about 10 wt. % methane, about 25 wt. % ethylene, and about 17 wt. % propylene, all wt. % being based on the total weight of said product, with the remainder consisting mostly of other hydrocarbon molecules having from 4 to 35 carbon atoms per molecule.

The cracked product is then further processed in the olefin production plant to produce, as products of the plant, various separate individual streams of high purity such as hydrogen, ethylene, propylene, mixed hydrocarbons having four carbon atoms per molecule, fuel oil, and pyrolysis gasoline. Each separate individual stream aforesaid is a valuable commercial product in its own right. Thus, an olefin production plant currently takes a part (fraction) of a whole crude oil stream and generates therefrom a plurality of separate, valuable products.

Natural gas and whole crude oil(s) were formed naturally in a number of subterranean geologic formations (formations) of widely varying porosities. Many of these formations were capped by impervious layers of rock. Natural gas and whole crude oil (crude oil) also accumulated in various stratigraphic traps below the earth's surface. Vast amounts of both natural gas and/or crude oil were thus collected to form hydrocarbon bearing formations at varying depths below the earth's surface. Much of this natural gas was in close physical contact with crude oil, and, therefore, absorbed a number of lighter molecules from the crude oil.

When a well bore is drilled into the earth and pierces one or more of such hydrocarbon bearing formations, natural gas and/or crude oil can be recovered through that well bore to the earth's surface.

The terms “whole crude oil” and “crude oil” as used herein means liquid (at normally prevailing conditions of temperature and pressure at the earth's surface) crude oil as it issues from a wellhead separate from any natural gas that may be present, and excepting any treatment such crude oil may receive to render it acceptable for transport to a crude oil refinery and/or conventional distillation in such a refinery. This treatment would include such steps as desalting. Thus, it is crude oil that is suitable for distillation or other fractionation in a refinery, but which has not undergone any such distillation or fractionation. It could include, but does not necessarily always include, non-boiling entities such as asphaltenes or tar. As such, it is difficult if not impossible to provide a boiling range for whole crude oil. Accordingly, whole crude oil could be one or more crude oils straight from an oil field pipeline and/or conventional crude oil storage facility, as availability dictates, without any prior fractionation thereof.

Natural gas, like crude oil, can vary widely in its composition as produced to the earth's surface, but generally contains a significant amount, most often a major amount, i.e., greater than about 50 weight percent (wt. %), methane. Natural gas often also carries minor amounts (less than about 50 wt. %), often less than about 20 wt. %, of one or more of ethane, propane, butane, nitrogen, carbon dioxide, hydrogen sulfide, and the like. Many, but not all, natural gas streams as produced from the earth can contain minor amounts (less than about 50 wt. %), often less than about 20 wt. %, of hydrocarbons having from 5 to 12, inclusive, carbon atoms per molecule (C5 to C12) that are not normally gaseous at generally prevailing ambient atmospheric conditions of temperature and pressure at the earth's surface, and that can condense out of the natural gas once it is produced to the earth's surface. All wt. % are based on the total weight of the natural gas stream in question.

When various natural gas streams are produced to the earth's surface, a hydrocarbon composition often naturally condenses out of the thus produced natural gas stream under the then prevailing conditions of temperature and pressure at the earth's surface where that stream is collected. There is thus produced a normally liquid hydrocarbonaceous condensate separate from the normally gaseous natural gas under the same prevailing conditions. The normally gaseous natural gas can contain methane, ethane, propane, and butane. The normally liquid hydrocarbon fraction that condenses from the produced natural gas stream is generally referred to as “condensate,” and generally contains molecules heavier than butane (C5 to about C20 or slightly higher). After separation from the produced natural gas, this liquid condensate fraction is processed separately from the remaining gaseous fraction that is normally referred to as natural gas.

Thus, condensate recovered from a natural gas stream as first produced to the earth's surface is not the exact same material, composition wise, as natural gas (primarily methane). Neither is it the same material, composition wise, as crude oil. Condensate occupies a niche between normally gaseous natural gas and normally liquid whole crude oil. Condensate contains hydrocarbons heavier than normally gaseous natural gas, and a narrow range of hydrocarbons that are at the lightest end of whole crude oil.

Condensate, unlike crude oil, can be characterized by way of its boiling point range. Condensates normally boil in the range of from about 100 to about 650 degrees Fahrenheit (F.). With this boiling range, condensates contain a wide variety of hydrocarbonaceous materials. These materials can include compounds that make up fractions that are commonly referred to as naphtha, kerosene, diesel fuel(s), and gas oil (fuel oil, furnace oil, heating oil, and the like). Naphtha and associated lighter boiling materials (naphtha) are in the C5 to C10, inclusive, range, and are the lightest boiling range fractions in condensate, boiling in the range of from about 100 to about 400 F. Petroleum distillates (kerosene, diesel, gas oil) are generally in the C10 to about C20 or slightly higher range, and generally boil, in their majority, in the range of from about 350 to about 650 F. They are, individually and collectively, referred to herein as “distillate” or “distillates.” It should be noted that various distillate compositions can have a boiling point lower than 350 F and/or higher than 650 F, and such distillates are included in the 350-650 F range aforesaid, and in this invention.

The starting feedstock for a conventional olefin production plant, as described above, normally has first been subjected to substantial, expensive processing before it reaches that plant. Normally, condensate and whole crude oil is distilled or otherwise fractionated into a plurality of fractions such as gasoline, naphtha, kerosene, gas oil (vacuum or atmospheric) and the like, including, in the case of crude oil and not natural gas, a high boiling residuum. Thereafter any of these fractions, other than the residuum, are normally passed to an olefin production plant as the starting feedstock for that plant.

It would be desirable to be able to forego the capital and operating cost of a refinery distillation unit that processes condensate and/or crude oil to generate a hydrocarbonaceous fraction that serves as the normal starting feedstock for conventional olefin producing plants. However, the prior art, until recently, taught away from even hydrocarbon cuts (fractions) that have too broad a boiling range distribution. For example, see U.S. Pat. No. 5,817,226 to Lenglet.

Recently, U.S. Pat. No. 6,743,961 (hereafter “USP '961”) issued to Donald H. Powers. This patent relates to cracking whole crude oil by employing a vaporization/mild cracking zone upstream of a cracking furnace. This zone is operated in a manner such that the liquid phase of the whole crude that has not already been vaporized is held in that zone until cracking/vaporization of the more tenacious hydrocarbon liquid components is maximized. This allows only a minimum of solid residue formation which residue remains behind as a deposit on the packing. This residue is later burned off the packing by conventional steam air decoking, ideally during the normal furnace decoking cycle. Thus, the second zone 9 of that patent serves as a trap for components, including hydrocarbonaceous materials, of the crude oil feed that cannot be cracked or vaporized under the conditions employed in the process.

U.S. Pat. No. 7,019,187 (hereafter “USP '187”) issued even more recently, and is directed to the process disclosed in USP '961, but employs a mildly acidic cracking catalyst to drive the overall function of the vaporization/mild cracking unit more toward the mild cracking end of the vaporization—mild cracking spectrum.

U.S. Pat. No. 6,979,757 (hereafter “USP '757”) is directed to the process disclosed in USP '961 but removes at least part of the liquid hydrocarbons remaining in the vaporization/mild cracking unit that are not yet vaporized or mildly cracked. These liquid hydrocarbon components of the crude oil feed are drawn from near the bottom of that unit and passed to a separate controlled cavitation device to provide additional cracking energy for those tenacious hydrocarbon components that have previously resisted vaporization and mild cracking.

USP '961, '187, and '757 (collectively “USP's”), the disclosures of which are incorporated herein in their entirety by reference, employ the vaporization/mild cracking unit in a thermal cracking plant upstream of a conventional cracking furnace(s). The use of this unit (“section 3” of the USP's) enables the cracking plant to accept liquid whole crude oil and/or liquid natural gas condensate materials, a wholly novel approach for the thermal cracking industry.

The novel use of a vaporization unit with or without a mild cracking function (vaporization/mild cracking unit) in a thermal cracking plant, imposed, particularly with mild cracking, severe operating conditions on that unit, and, therefore, severe material construction restraints for that unit. By virtue of the mild cracking function, this unit (section 3 of the USP's) contains very high temperature steam, and is, in essence, a pressure vessel that, for extended periods of time during operation, must hold, at elevated temperatures and pressures, large volumes not only of steam, but also of the liquid feed aforesaid.

As disclosed in greater detail hereinafter, the particular material used in the construction of such a unit required an equally novel approach that was not at all obvious to one skilled in the art. This invention provides just such an approach.


In accordance with this invention, the vaporization/mild cracking unit employed in the thermal cracking process exemplified by the USP's is constructed of a martensitic steel containing chromium, molybdenum, and vanadium.

This class of steel was not heretofore used in pressure vessel applications, its use up to the present being confined to pipe and tubing applications.

Further in accordance with this invention, the as constructed unit is subjected to a de-hydrogenation treatment.


The terms “hydrocarbon,” “hydrocarbons,” and “hydrocarbonaceous” as used herein do not mean materials strictly or only containing hydrogen atoms and carbon atoms. Such terms include materials that are hydrocarbonaceous in nature in that they primarily or essentially are composed of hydrogen and carbon atoms, but can contain other elements such as oxygen, sulfur, nitrogen, metals, inorganic salts, and the like, even in significant amounts.

The term “gaseous” as used in this invention means one or more gases in an essentially vaporous state, for example, steam alone, a mixture of steam and hydrocarbon vapor, and the like.

The term “coke” as used in this invention means any high molecular weight carbonaceous solid, and includes compounds formed from the condensation of polynuclear aromatics.

An olefin producing plant useful with this invention would include a pyrolysis (thermal cracking) furnace for initially receiving and cracking the feed. Pyrolysis furnaces for steam cracking of hydrocarbons heat by means of convection and radiation, and comprise a series of preheating, circulation, and cracking tubes, usually bundles of such tubes, for preheating, transporting, and cracking the hydrocarbon feed. The high cracking heat is supplied by burners disposed in the radiant section (sometimes called “radiation section”) of the furnace. The waste gas from these burners is circulated through the convection section of the furnace to provide the heat necessary for preheating the incoming hydrocarbon feed. The convection and radiant sections of the furnace are joined at the “cross-over,” and the tubes referred to hereinabove carry the hydrocarbon feed from the interior of one section to the interior of the next.

Cracking furnaces are designed for rapid heating in the radiant section starting at the radiant tube (coil) inlet where reaction velocity constants are low because of low temperature. Most of the heat transferred simply raises the hydrocarbons from the inlet temperature to the reaction temperature. In the middle of the coil, the rate of temperature rise is lower but the cracking rates are appreciable. At the coil outlet, the rate of temperature rise increases somewhat but not as rapidly as at the inlet. The rate of disappearance of the reactant is the product of its reaction velocity constant times its localized concentration. At the end of the coil, reactant concentration is low and additional cracking can be obtained by increasing the process gas temperature.

Steam dilution of the feed hydrocarbon lowers the hydrocarbon partial pressure, enhances olefin formation, and reduces any tendency toward coke formation in the radiant tubes.

Cracking furnaces typically have rectangular fireboxes with upright tubes centrally located between radiant refractory walls. The tubes are supported from their top.

Firing of the radiant section is accomplished with wall or floor mounted burners or a combination of both using gaseous or combined gaseous/liquid fuels. Fireboxes are typically under slight negative pressure, most often with upward flow of flue gas. Flue gas flow into the convection section is established by at least one of natural draft or induced draft fans.

Radiant coils are usually hung in a single plane down the center of the fire box. They can be nested in a single plane or placed parallel in a staggered, double-row tube arrangement. Heat transfer from the burners to the radiant tubes occurs largely by radiation, hence the thermo “radiant section,” where the hydrocarbons are heated to from about 1,450° F. to about 1,550° F. and thereby subjected to severe cracking.

The initially empty radiant coil is, therefore, a fired tubular chemical reactor. Hydrocarbon feed to the furnace is preheated to from about 900° F. to about 1,000° F. in the convection section by convectional heating from the flue gas from the radiant section, steam dilution of the feed in the convection section, or the like. After preheating, in a conventional commercial furnace, the feed is ready for entry into the radiant section.

In a typical furnace, the convection section can contain multiple zones. For example, the feed can be initially preheated in a first upper zone, boiler feed water heated in a second zone, mixed feed and steam heated in a third zone, steam superheated in a fourth zone, and the final feed/steam mixture preheated to completion in the bottom, fifth zone. The number of zones and their functions can vary considerably. Thus, pyrolysis furnaces can be complex and variable structures.

The cracked gaseous hydrocarbons leaving the radiant section are rapidly reduced in temperature to prevent destruction of the cracking pattern. Cooling of the cracked gases before further processing of same downstream in the olefin production plant recovers a large amount of energy as high pressure steam for re-use in the furnace and/or olefin plant. This is often accomplished with the use of transfer-line exchangers that are well known in the art.

Downstream processing of the cracked hydrocarbons issuing from the furnace varies considerably, and particularly based on whether the initial hydrocarbon feed was a gas or a liquid. Since this invention applies to the use of liquid crude oil and/or natural gas condensate as a feed, downstream processing herein will be described for a liquid fed olefin plant. Downstream processing of cracked gaseous hydrocarbons from liquid feedstock is more complex than for gaseous feedstock because of the heavier hydrocarbon components present in the liquid feedstock.

With a liquid hydrocarbon feedstock downstream processing, although it can vary from plant to plant, typically employs an oil quench of the furnace effluent after heat exchange of same in, for example, the transfer-line exchanger aforesaid. Thereafter, the cracked hydrocarbon stream is subjected to primary fractionation to remove heavy liquids, followed by compression of uncondensed hydrocarbons, and acid gas and water removal. Various desired products are then individually separated, e.g., ethylene, propylene, a mixture of hydrocarbons having four carbon atoms per molecule, fuel oil, pyrolysis gasoline, and a high purity hydrogen stream.

The vaporization/mild cracking unit of this invention receives the liquid feed that may or may not have been preheated, for example, from about ambient to about 350 F, preferably from about 200 to about 350 F. This is a lower temperature range than what is required for complete vaporization of the feed. Any preheating preferably, though not necessarily, takes place in the convection section of the same furnace for which such condensate is the primary feed.

As shown in the USP's, the first zone in the vaporization/mild cracking operation employs vapor/liquid separation wherein vaporous hydrocarbons and other gases, if any, in the preheated feed stream are separated from those distillate components that remain liquid after preheating. The aforesaid gases are removed from the vapor/liquid separation section and passed on to the radiant section of the cracking furnace.

Vapor/liquid separation in this first, e.g., upper, zone knocks out distillate liquid in any conventional manner, numerous ways and means of which are well known and obvious in the art. Suitable devices for liquid vapor/liquid separation include liquid knock out vessels with tangential vapor entry, centrifugal separators, conventional cyclone separators, schoepentoeters, vane droplet separators, and the like.

Liquid thus separated from the aforesaid vapors moves into a second, e.g., lower, zone, as shown in the USP's. The liquid entering and traveling along the length of this second zone meets oncoming, e.g., rising, steam. This liquid, absent the removed gases, receives the full impact of the oncoming steam's thermal energy and diluting effect.

This second zone can carry at least one liquid distribution device such as a perforated plate(s), trough distributor, dual flow tray(s), chimney tray(s), spray nozzle(s), and the like.

This second zone can also carry in a portion thereof one or more conventional tower packing materials and/or trays for promoting intimate mixing of liquid and vapor in the second zone.

As the remaining liquid hydrocarbon travels (falls) through this second zone, lighter materials such as gasoline or naphtha that may be present can be vaporized in substantial part by the high energy steam with which it comes into contact. This enables the hydrocarbon components that are more difficult to vaporize to continue to fall and be subjected to higher and higher steam to liquid hydrocarbon ratios and temperatures to enable them to be vaporized by both the energy of the steam and the decreased liquid hydrocarbon partial pressure with increased steam partial pressure.

The elevated temperature and pressure conditions under which the vaporization/mild cracking pressure vessel unit (pressure vessel unit) operates are quite rigorous, particularly as to the elevated temperatures that are to be maintained within and throughout this pressure vessel unit. As disclosed in the USP's, the steam entering the bottom of the pressure vessel unit (section 3 of the USP's) is at a temperature of from about 1,000 to about 1,300 F in order to maintain a mild cracking temperature inside that unit of from about 800 to about 1,300 F. Thus, at least the bottom portion of that unit, and even the entire mild cracking zone of that unit, can, in operation, consistently be at or above 1,200 F. The upper portions of that unit can consistently be at an operating temperature of at least about 900 F. While operating at these elevated temperatures, the pressure vessel unit is also under a pressure of at least about 50 psig.

In addition, due to the liquid feed to this pressure vessel unit, the material from which the pressure vessel unit itself is constructed will be exposed to sulfur and/or sulfur containing compounds.

Further, due to routine maintenance and de-coking requirements, the pressure vessel unit is exposed on a periodic basis to ambient atmospheric conditions, including water vapor and oxygen. For example, the pressure vessel unit could be cooled from 1,200 F to 600 F on a regular basis for de-coking purposes, and, at longer time intervals, cooled from 1,200 F to ambient atmospheric conditions for maintenance purposes.

Heretofore, in pressure vessel construction for high temperature, elevated pressure applications, the prior art has employed refractory lined vessels or fabricated such vessels from ferritic steel or austenitic stainless steel.

In the vaporization/mild cracking context described hereinabove a refractory lined vessel is completely unacceptable. First of all the presence of the refractory itself in the interior of the vessel would, as is well known in the art, promote severe coke deposition on the refractory. Further, the liquid hydrocarbonaceous feeds employed in the process of this invention would, on first contact, rapidly degrade the refractory.

A pressure vessel composed of iron based ferritic steel such as steel containing chromium and molybdenum is also not useful in the foregoing mild cracking context because such steels have a maximum operating temperature of about 1,000 F. At 1,200 F ferritic steels have inadequate strength for a prolonged thermal cracking operation. This inadequate strength factor could arguably be compensated for by employing thicker wall members, but then thermal cycling fatigue due to the maintenance and de-coking operations aforesaid would become a substantial and constant problem.

Iron based austenitic stainless steels such as the 300 series of chrome/nickel stainless steels has a high coefficient of expansion and low thermal conductivity. Under the mild cracking/maintenance/de-coking regime aforesaid a vessel constructed of this type steel would have unacceptably low resistance to thermal fatigue. Just as importantly, when exposed to elevated temperatures of from about 700 to about 1,400 F for extended time periods this type of steel tends to form carbide precipitate at its grain boundaries which sensitizes the steel itself. Upon subsequent exposure of the sensitized steel to a combination of oxygen, water vapor, sulfur/sulfur compounds, and sulfide scale, this type of steel can develop inter-granular corrosion (without cracking) and inter-granular stress corrosion cracking in an aqueous atmosphere. If formed from austenitic stainless steel, the lower portion of the pressure vessel unit, e.g., second zone 9 of the USP' would be particularly susceptible to inter-granular stress corrosion cracking due to its constant exposure to at least 1,000 F steam. Such cracking was heretofore observed when austenitic stainless steel was employed in the tubes of a conventional thermal cracking furnace.

Pursuant to this invention, the pressure vessel unit of the USP's is constructed primarily of an iron based martensitic steel containing minor amounts (less than about 10 wt. % each based on the total weight of the alloy) of chromium, molybdenum, and vanadium.

This class of steel has both a lower coefficient of thermal expansion and a higher thermal conductivity than austenitic stainless steel.

This class of steel also has no susceptibility to inter-granular corrosion or stress corrosion cracking.

This steel also has good resistance to corrosion caused by sulfidation mechanisms.

Finally, this steel has greater strength than ferritic steels which allows for the pressure vessel unit to have thinner walled heads and shell. This improves substantially the unit's resistance to thermal fatigue.

Particularly suitable martensitic steels useful in this invention are Grade 91 steel—ASME Section 2A SA387 which nominally contains about 9 wt. % chromium, about 1 wt. % molybdenum, and about 0.2 wt. % vanadium, all wt. % based on the total weight of the steel, and related alloys such as P92 and E911 steels which are well known in the art and fully and completely disclosed in “Recent Advances in Creep-Resistant Steels for Power Plant Applications” by P. J. Ennis and A. Czyrska-Filemonowicz, and published in “Sadhana,” Vol. 28, Parts 3 & 4, June/August 2003, pp. 709-730.

This type of martensitic steel is commercially available for pipe and tubing fabricators, and, once appraised of this invention, can be obtained commercially by a fabricator for use in the practice of this invention.

Heretofore, martensitic steel has not been used in pressure vessel applications in general because of concern for hydrogen cracking and “as welded” fracture toughness. These concerns have been mitigated heretofore when making pipe or tubing from this type of steel by employing an electric heat wrap on each weld promptly after the formation of that weld. This is not feasible for large pressure vessels that can, in the example of the pressure vessel units of this invention be 5 feet in diameter and 60 feet long.

However, it has been found, pursuant to this invention, that martensitic steel can be employed in the context of a vaporization/mild cracking pressure vessel application, the foregoing hydrogen cracking and as welded fracture toughness concerns notwithstanding. Pursuant to this invention the pressure vessel unit is formed from the foregoing martensitic steel by using conventional welding procedures well known in the art. Thereafter, the thus constructed vessel, in its entirety, is subjected to a de-hydrogenation treatment. This heat treatment can vary widely depending on the size of the pressure vessel and the particular welding procedures and other construction techniques employed in fabricating the particular vessel in question. However, the post-welding heat treatment of this invention, as applied to the completed vessel, will generally involve heating the vessel as a whole at a temperature of at least about 600 F for at least about 2 hours under ambient atmosphere. By heat treating the vessel as constructed, the concerns that have heretofore kept the prior art from using martensitic steels in the fabrication of pressure vessels are overcome.