WROUGHT, NICKEL-BASED ALLOY WITH ENHANCED RESISTANCE TO WET-PROCESS PHOSPHORIC ACID AND CREVICE CORROSION

FIELD OF INVENTION

[0001]The invention relates to nickel-based alloys which are resistant to wet-process phosphoric acid and crevice corrosion.

BACKGROUND

[0002]“Wet process” phosphoric acid (P₂O₅), which is made by reacting phosphate rock with sulfuric acid, is one of the most important industrial chemicals, being the primary source of phosphorus for agrichemical fertilizers. As produced, it contains many impurities, and has a P₂O₅ concentration of only about 30%, because of the large amount of rinse water needed to separate it from the other main reaction product, calcium sulfate. Typical impurities include unreacted sulfuric acid, various metallic ions, fluoride ions, and chloride ions. The fluoride ions tend to form complexes with the metallic ions, and are therefore less of a problem than the chloride ions, which strongly influence electrochemical reactions between “wet process” phosphoric acid and metallic materials. Particulate matter (for example, silica particles) can also be present in “wet process” acid.

[0003]The main use of metallic materials is in the concentration process, where the “wet process” acid is taken through a series of evaporation steps, involving metallic tubing. Typically, the P₂O₅ concentration is raised to 54% during this process. To help with transportation, additional concentration/evaporation steps can be added. However, the corrosivity of P₂O₅ at concentrations above 54% can be lower, due to reduced impurity levels.

[0004]In the early 1980's, a new, wrought, nickel-based material, HASTELLOY G-30 alloy (U.S. Pat. No. 4,410,489), was designed specifically for use in “wet process” phosphoric acid evaporator tubes. It had a high chromium content (about 30 wt. %), to cope with the ferric ion impurities known to be present in P₂O₅, and a combination of molybdenum plus tungsten to provide resistance to chloride-induced, localized attack (in particular “under-deposit” corrosion, which is similar to crevice corrosion and is known to be a problem in P₂O₅ evaporator tubes). As in a prior wrought alloy (designated HASTELLOY G-3 alloy) used for this industrial application, a copper content of 2 wt. % was included in HASTELLOY G-30 alloy, presumably to help resistance to impurity levels of unreacted sulfuric acid. Like HASTELLOY G-3 alloy, it had a significant iron content (15 wt. %) and therefore fell into the nickel-chromium-iron (Ni—Cr—Fe) category of nickel-based alloys.

[0005]In the early 2000's, HASTELLOY G-30 alloy was superseded by a wrought, nickel-chromium-molybdenum (Ni—Cr—Mo) material (HASTELLOY G-35 alloy, U.S. Pat. No. 6,740,291) with a higher chromium content and low iron content. Studies during that time indicated that copper was not necessary for high resistance to “wet process” phosphoric acid, and might be a de-stabilizer of the microstructure of such alloys. Also, it was determined that a chromium aim of about 33 wt. % and a molybdenum aim of about 8 wt. % were not only amenable to carbon arc furnace/argon oxygen de-carburization (CAF/AOD) melting, but also resulted in even higher resistance to P₂O₅ and chloride-induced crevice attack.

[0006]Although HASTELLOY G-35 alloy is suitable for use for many applications in the chemical industry that involve wet process phosphoric acid, there is a need for an alloy which has enhanced resistance to P₂O₅, and equal or greater resistance to chloride-induced crevice corrosion relative to HASTELLOY G-35 alloy.

SUMMARY OF THE INVENTION

[0007]We provide a nickel based alloy which contains 38.10 to 38.21 wt. % chromium, 6.90 to 8.11 wt. % molybdenum, 0.81 to 0.83 wt. % iron, 0.19 to 0.20 wt. % aluminum, 0.27 to 0.29 wt. % manganese, 0.184 to 0.217 wt. % nitrogen, and the balance nickel plus impurities, including silicon and carbon. This alloy corrodes at rates which are at least 25% lower than HASTELLOY G-35 alloy in 54% and 68% “wet-process” phosphoric acid at 121° C. and possesses higher resistance to chloride-induced crevice attack.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008]The purpose of the work that led to this discovery was to investigate the effects of even higher chromium contents than in HASTELLOY G-35 alloy, along with other elemental changes, in the hope of enhanced resistance to P₂O₅ and equal or enhanced resistance to chloride-induced crevice corrosion, while maintaining hot process-ability, to enable the production of wrought forms, such as plates, sheets, bars, and wires.

[0009]The aim and actual compositions of the wrought alloys studied during the work are given in Table 1. Apart from Alloy A, the composition of which was designed to replicate HASTELLOY G-35 alloy to enable more precise data comparisons, the alloys studied covered an approximate chromium content range of 36 to 40 wt. %, and an approximate molybdenum content range of 4 to 8 wt. %. Nitrogen was also added to several of the experimental alloys in the approximate range 0.15 to 0.25 wt. %, given that it is known to enhance the resistance to chloride-induced crevice attack of stainless steels, if soluble.

[0010]The precise solubility of nitrogen in nickel-based alloys is unknown. However, it is believed that the higher the chromium content, the higher is the nitrogen solubility, based upon the levels of nitrogen absorbed into various nickel-based alloys during air-melting.

[0011]The elements iron, aluminum, and manganese were added to all of the experimental alloys at (constant) levels typical of wrought, corrosion-resistant, nickel-based alloys without deliberate iron additions, such as HASTELLOY G-35 alloy. In other words, it is typical to encounter impurity iron contents close to 0.75 wt. % in alloys such as HASTELLOY G-35 alloy. Aluminum contents of approximately 0.25 wt. % are typical of wrought, corrosion-resistant, nickel-based alloys, since aluminum is used to remove oxygen from alloys in the molten state, and is used to maintain molten material temperatures during argon-oxygen de-carburization (AOD). Likewise, manganese contents of approximately 0.25 wt. % (or greater) are used to remove sulfur from such alloys in the molten state.

[0012]As is usual for wrought, corrosion-resistant, nickel-based alloys, the contents of both silicon and carbon were kept as low as possible in the experimental materials. This is because silicon can promote the formation of deleterious, second-phase precipitates in nickel-based alloys at temperatures in excess of about 1000° F. (i.e. during cool-down from the molten state, or during elevated temperature excursions, as might occur during welding). Carbon can cause the formation of deleterious grain boundary carbide precipitates at temperatures in excess of about 1000° F., where elemental diffusion becomes significant.

[0013]It should be mentioned that most wrought, corrosion-resistant, nickel-based alloys are thermally unstable (i.e. supersaturated with alloying elements, in order to maximize their corrosion resistance). This can be overcome by a heat treatment known as solution annealing (typically at temperatures of 1950° F. and above), followed by rapid quenching (generally with cold water).

[0014]The solution annealing temperature is one at which the alloying additions are completely soluble in nickel, forming a face-centered cubic (FCC, or gamma phase) atomic structure. Upon rapid quenching to room temperature, this atomic structure (albeit in a metastable state) is locked in.

[0015]The operating temperatures of such alloys are generally below 500° F., so there is little concern over changes to the atomic and metallurgical structures in service. However, during welding, so-called heat-affected zones (HAZ's) adjacent to the molten weld pool can encounter thermal cycles involving temperatures well above 1000° F.

[0016]The wrought, experimental alloys involved with this study were made by vacuum induction melting (VIM), followed by electro-slag re-melting (ESR), to produce ingots of material (of diameter 4 in, length 7 in, and approximate weight 25 lb.) amenable to hot working. Ingots of these experimental alloys were homogenized for 24 hours prior to hot forging, in the temperature range 2175° F. to 2225° F. (the typical range for HASTELLOY G-35 alloy). Hot forging, and subsequently hot rolling, of these alloys were carried out using start temperatures within the same range (again, typical of HASTELLOY G-35 alloy). The alloys were hot rolled to plates, of thickness 0.5 in, and sheets, of thickness 0.125 in. The successful production of wrought plates and sheets during this study indicates that a homogenization, forging start, and rolling start temperature of about 2200° F. is suitable for these alloys.

[0017]Trials were performed to determine an appropriate solution annealing temperature for each alloy, i.e. one that resulted in structures free of remnants of the hot worked microstructure, free of second-phase precipitates, and with moderate grain sizes (ASTM 3 to 5). In most cases, this was 2100° F. The trials involved heating the alloys for 20 minutes at various temperatures in the range 2000° F. to 2225° F., followed by water quenching, sectioning, polishing, etching, and metallographic examination using optical microscopes.

[0021]Two industrial batches of P₂O₅ were used in this study. The concentration of one batch was 54%, i.e. the concentration typically derived from the first series of evaporation steps, and generally the most corrosive concentration. The concentration of the other batch was 68%, i.e. after additional evaporation steps, to help with transportation. Testing was performed at 121° C. in an autoclave, using an uninterrupted duration of 96 hours. 121° C. is a typical operating temperature during the evaporation steps.

[0022]The results of testing the experimental materials in P₂O₅, alongside the relevant commercial alloys (for comparison) are given in Table 2. From these results, it was deduced that the resistance to P₂O₅ of high-chromium, molybdenum-bearing, nickel-based alloys is not purely a function of the chromium content, but is also influenced by the molybdenum content, and whether or not nitrogen is deliberately added.

[0023]As mentioned previously, nitrogen was added to several experimental alloys in an attempt to improve their crevice corrosion resistance. Its positive effect upon P₂O₅ resistance (deduced by comparing the results for Alloys E and F) was unexpected and unpredicted. To our knowledge, nitrogen has not been added to nickel-based alloys with such high chromium contents before.

[0024]From the test results shown in Table 2, it is evident that Alloy F provides the highest resistance to both 54% and 68% P₂O₅. Compared with Alloy A (an experimental version of HASTELLOY G-35 alloy), Alloy F provides a 23% lower corrosion rate in 54% P₂O₅, and a 37% lower corrosion rate in 68% P₂O₅. Compared with commercial HASTELLOY G-35 alloy, Alloy F provides a 29% lower corrosion rate in 54% P₂O₅, and a 33% lower corrosion rate in 68% P₂O₅.

[0025]Alloys G, H, and I, also exhibited high resistance to both concentrations of P₂O₅, especially relative to HASTELLOY G-35 alloy. All three of these alloys provided a 25% lower corrosion rate than HASTELLOY G-35 alloy in 54% P₂O₅, and in 68% P₂O₅, Alloy G was equal to Alloy F (with a 33% lower corrosion rate relative to HASTELLOY G-35 alloy), while Alloys H and I provided a 31% lower corrosion rate relative to HASTELLOY G-35 alloy.

[0026]To assess the resistance of the experimental materials to chloride-induced crevice corrosion, they were subjected to the test described in ASTM Standard G48, Method D. This test involves the attachment of crevice assemblies to sheet samples, then submersion of these samples in 6% ferric chloride+1% hydrochloric acid for an uninterrupted duration of 72 hours. Tests are usually performed at several temperatures to enable a determination of the critical crevice temperature (CCT). The CCT of a material is the lowest temperature at which crevice attack is observed (on either sample, or on both samples, in this case). Two samples of each material were tested at each temperature in this study.

[0027]The results of testing the experimental alloys according to the procedures defined in ASTM Standard G48, Method D are shown in Table 3. The corresponding CCT's are given in Table 4, along with those previously determined for the commercial materials, HASTELLOY G-30 alloy and HASTELLOY G-35 alloy. The corrosion rates shown in Table 3 represent averages over the whole surface of each sample (i.e. those areas covered by the crevice assembly, and those not covered). Since the uncovered areas generally corrode very little, the values for those samples which suffered crevice attack provide some information on the depth of attack, or the number of attack sites.

[0028]Comparing the CCT values for Alloys E (with no deliberate nitrogen addition) and F (containing 0.184 wt. % nitrogen), the data confirm that nitrogen is very beneficial to chloride-induced crevice corrosion resistance. It also suggests that nitrogen is soluble at the 0.184 wt. % level in these high-chromium, nickel-based alloys. Comparing the CCT values for Alloys H (with a molybdenum content of 8.11 wt. %) and I (with a molybdenum content of 5.85 wt. %), it is evident that molybdenum also has a strong positive effect upon resistance to crevice corrosion, as expected. It should be highlighted that the actual molybdenum content of Alloy I (5.85 wt. %) was well below its aim molybdenum content (8.0 wt. %); this was probably due to a furnace charge miscalculation or weighing error, but did provide a useful comparison.

[0029]Since the CCT of Alloy I (40° C.) was less than that of both Alloy A and HASTELLOY G-35 alloy (both 45° C.), Alloy I did not meet the objectives of the study, which were enhanced resistance to P₂O₅, and equal or greater resistance to chloride-induced crevice corrosion relative to Alloy A and HASTELLOY G-35 alloy. On the other hand, Alloys F, G, and H met the objectives, and provided considerably higher crevice corrosion resistance.

[0030]The results in Tables 2, 3, and 4 indicate that nitrogen is key to attaining the highest resistance to both P₂O₅ and to chloride-induced crevice attack in high-chromium, molybdenum-bearing, nickel-based alloys. Furthermore, it was unexpectedly discovered that nitrogen also alters the nature of the second phases that occur during aging in the approximate temperature range 1400° F. to 1700° F. If Alloy E, without a deliberate addition of nitrogen, is held within this temperature range for 25 hours, then it is prone to the secondary (i.e. diffusion induced) precipitation of an acicular (needle-like) intermetallic phase, believed to be the well-known sigma phase.

[0031]If Alloy F, containing 0.184 wt. % nitrogen, is held within this temperature range for 25 hours, then it is prone to a eutectoid reaction, during which the gamma solid solution transforms to a layered (lamellar) structure, believed to comprise layers of gamma phase and alpha chromium.

[0032]While such changes at high temperature are not encountered in evaporators for the concentration of P₂O₅, tensile tests indicate that the lamellar structure might be more ductile than a structure containing needle-like precipitates, thus benefitting hot process-ability in the production of wrought products.

[0033]During the manufacture of nickel-based alloys, it is not possible to attain precise control of elemental contents to two decimal places, given that variances can occur in chemical analyses, and during the melting processes. Ranges are therefore established, based on prior knowledge of such variances. Any commercial embodiment of the alloys of this invention (including the iron, aluminum, and manganese additions, and silicon and carbon impurity levels) would therefore be subject to such typical manufacturing variances.

[0040]The variance expected for nitrogen in HAYNES HR-120 alloy, a high temperature, iron-nickel-chromium material containing a similar nitrogen content (0.21 wt. %) to that of Alloy F, is plus or minus 0.03 wt. %

[0065]In addition, silicon and carbon should be kept as low as possible in the commercial embodiment of Alloys F, G, and H, but might be as high as 0.050 wt. % (in the case of silicon) and 0.015 (in the case of carbon), according to the expectations for HASTELLOY G-35 alloy (which also calls for silicon and carbon to be as low as possible).

[0075]Although we have described certain present preferred embodiments of our alloy it should be understood that our invention is not limited thereto but may be variously embodied within the scope of the following claims.