STEAM REFORMING

Abstract

A process is described for steam reforming a hydrocarbon feedstock containing one or more nitrogen compounds, comprising passing a mixture of the hydrocarbon feedstock and steam through a catalyst bed consisting of one nickel steam reforming catalysts disposed within a plurality of externally heated tubes in a tubular steam reformer, wherein each tube has an inlet to which the mixture of hydrocarbon and steam is fed, an outlet from which a reformed gas containing hydrogen, carbon monoxide, carbon dioxide, steam, ammonia and methane is recovered, and the steam reforming catalyst at least at the outlet of the tubes is a particulate eggshell steam reforming catalyst comprising 2.5 to 9.5% by weight nickel, expressed as NiO, wherein the nickel is provided in a layer at the surface of the catalyst and the thickness of layer is in the range of 100 to 1000 μm.

Claims

1-19. (canceled)

20. A process for steam reforming a hydrocarbon feedstock containing one or more nitrogen compounds, comprising passing a mixture of the hydrocarbon feedstock and steam through a catalyst bed consisting of one or more nickel steam reforming catalysts disposed within a plurality of externally heated tubes in a tubular steam reformer, wherein each tube has an inlet to which the mixture of hydrocarbon and steam is fed, an outlet from which a reformed gas containing hydrogen, carbon monoxide, carbon dioxide, steam, ammonia and methane is recovered, and the steam reforming catalyst at least at the outlet of the tubes is a particulate eggshell steam reforming catalyst comprising 2.5 to 9.5% by weight nickel, expressed as NiO, wherein the nickel is provided in a layer at the surface of the catalyst and the thickness of layer is in the range of 100 to 1000 μm.

21. The process according to claim 20, wherein the nickel is provided in a layer at the surface of the catalyst and the thickness of layer is in the range of 100 to 800 μm.

22. The process according to claim 20, wherein the nickel is supported on a shaped particulate catalyst support comprising alumina, titania or zirconia or an alkaline earth metal aluminate.

23. The process according to claim 20, wherein the nickel content of the eggshell catalyst, expressed as NiO, is in the range 2.5 to 5.5% by weight.

24. The process according to claim 20, wherein the hydrocarbon feedstock comprises methane, a pre-reformed gas, an associated gas or natural gas.

25. The process according to claim 20, wherein the hydrocarbon feedstock is compressed to a pressure in the range 10 to 100 bar abs.

26. The process according claim 20, wherein the one or more nitrogen compounds comprises nitrogen gas, N2.

27. The process according to claim 26, wherein the nitrogen gas content of the hydrocarbon feedstock is in the range of 0.1 to 25% by volume.

28. The process according to claim 20, wherein the mixture of hydrocarbon feedstock and steam has a steam to carbon ratio in the range 1.8:1 to 5:1.

29. The process according to claim 20, wherein the mixture of hydrocarbon feedstock and steam is fed to the inlets of the tubes at an inlet temperature in the range 300 to 650° C.

30. The process according to claim 20, wherein the tubular steam reformer contains a plurality of tubes through which the mixture of the hydrocarbon feedstock and steam is passed, and to which heat is transferred by means of a hot gas comprising a combustion gas or a synthesis gas, flowing around the tubes.

31. The process according to claim 20, wherein the catalyst bed consists of one, two, three or more layers of steam reforming catalyst wherein in each case the layer of steam reforming catalyst adjacent the outlets of the tubes is the eggshell nickel catalyst.

32. The process according to claim 31, wherein there are two or more layers of steam reforming catalyst within the tubes and the eggshell catalyst layer comprises 95% to 5% of the volume of the bed.

33. The process according to claim 20, wherein the methane content of the reformed gas is less than 15% by volume on a dry gas basis.

34. The process according to claim 20, wherein the ammonia content of the reformed gas is below 200 ppmv on a dry gas basis.

35. The process according to claim 20, wherein the process further comprises cooling the reformed gas to below the dew point to condense steam and separating the liquid condensate to form a synthesis gas from the reformed gas.

36. The process according to claim 35, wherein the ammonia content of the liquid condensate is below 400 mg/Litre.

37. The process according to claim 35, wherein at least a portion of the condensate is recycled and used to generate steam used in the steam reforming process.

38. The process according to claim 26, wherein the nitrogen gas content of the hydrocarbon feedstock is in the range of 0.5 to 25% by volume.

39. The process according to claim 26, wherein the nitrogen gas content of the hydrocarbon feedstock is in the range of 1 to 10% by volume.

40. The process according to claim 31, wherein there are two or more layers of steam reforming catalyst within the tubes and the eggshell catalyst layer comprises 80% to 20% of the volume of the bed.

41. The process according to claim 31, wherein there are two or more layers of steam reforming catalyst within the tubes and the eggshell catalyst layer comprises 75% to 25% of the volume of the bed.

42. The process according to claim 20, wherein the methane content of the reformed gas is less than 10% by volume on a dry gas basis.

43. The process according to claim 20, wherein the methane content of the reformed gas is less than 5% by volume on a dry gas basis.

44. The process according to claim 20, wherein the ammonia content of the reformed gas is below 100 ppmv on a dry gas basis.

45. The process according to claim 20, wherein the ammonia content of the reformed gas is below 50 ppmv on a dry gas basis.

46. The process according to claim 20, wherein the ammonia content of the reformed gas is below 10 ppmv on a dry gas basis.

47. The process according to claim 35, wherein the ammonia content of the liquid condensate is below 200 mg/Litre.

48. The process according to claim 35, wherein the ammonia content of the liquid condensate is below 100 mg/Litre.

49. The process according to claim 35, wherein the ammonia content of the liquid condensate is below 20 mg/Litre.

Description

[0044] The invention is further described by reference to the following Examples and FIGS. 1 to 6, in which:

[0045] FIG. 1 is a graph depicting the ammonia produced per second per gram Ni versus bed temperature for catalysts in tests using a reformer feed containing 2% vol N.sub.2;

[0046] FIG. 2 is a graph depicting the ammonia produced per second per gram Ni versus bed temperature for catalysts in tests using a reformer feed containing 5% vol N.sub.2;

[0047] FIG. 3 is a graph depicting the ammonia produced per second per gram Ni versus bed temperature for catalysts in tests using a reformer feed containing 8% vol N.sub.2;

[0048] FIG. 4 is a graph depicting the ammonia produced per second versus % mol ethane conversion for catalysts in tests using a reformer feed containing 2% vol N.sub.2;

[0049] FIG. 5 is a graph depicting the ammonia produced per second versus % mol ethane conversion for catalysts in tests using a reformer feed containing 5% vol N.sub.2; and

[0050] FIG. 6 is a graph depicting the ammonia produced per second versus % mol ethane conversion for catalysts in tests using a reformer feed containing 8% vol N.sub.2.

EXAMPLE 1

[0051] Tests were performed with various nickel catalysts for the steam reforming of natural gas containing 1.5% vol N.sub.2. Steam and nitrogen gas were combined with the natural gas such that the reforming feed gas mixture fed to the catalysts comprised 2, 5 or 8% vol N.sub.2 on a wet gas basis.

[0052] The following pelleted steam reforming catalysts were tested:

TABLE-US-00001 Catalytic NiO Catalyst Shape metal Support Type content Amount 1(a) 3.3 × 3.3 mm Ni Ca- Non- 17.6% wt  24.94 g Comparative cylinder aluminate eggshell 1(b) 3.3 × 3.3 mm Ni Ca- Non- 7.2% wt 22.85 g Comparative cylinder aluminate eggshell 1(c) 3.3 × 3.3 mm Ni Ca- Eggshell 5.0% wt 22.50 g Eggshell cylinder aluminate

[0053] The non-eggshell catalysts have the nickel distributed evenly though the pellet. In the eggshell catalyst, the nickel is present only in a layer approximately 400 micrometres thick at the surface of the pellet.

[0054] The eggshell nickel catalyst was prepared according to the eggshell catalyst method of WO2010/125369. The non-eggshell catalysts were prepared according to the comparative examples in WO2010/125369.

[0055] The catalysts were tested in a laboratory scale steam reformer with a single electrically heated reformer tube with an internal diameter of about 25 mm and a length of about 2100 mm. The reactor operated on an up-flow basis. Water for generating steam was supplied to the rig via a variable stroke pump and was fed to the bottom of the reactor where it was vaporised. Natural gas was fed through a separate desulphurisation vessel before being delivered to the reactor via a thermal mass flow controller. Nitrogen and hydrogen were also be supplied to the reactor via independent mass flow controllers if required. The water and gases all entered the reactor via the same inlet pipe. The product gas exited the reactor via an outlet from the tube and was cooled to ambient temperature to condense the steam which was then collected in a catch-pot. A small volume of dry exit gas was fed to a Varian CP490 quad-channel micro GC analyser. This gas then returned to the exit gas meter to allow for a full mass balance from the reformer to be calculated.

[0056] The catalysts were diluted to approximately 100 ml with fused alumina chips (sieve fraction 3.35 mm-4.74 mm) and installed as a layer near the outlet of the reformer tube. The remainder of the tube was charged with 3.35-4.75 mm alumina chips. At the start of each test the catalyst was reduced using 50 vol % H.sub.2 in N.sub.2 at 600° C. for 2 hours.

[0057] Reforming was then carried out at a pressure of 27 barg using bed inlet temperatures in the range of 610 to 800° C. with a steam to carbon ratio of 3:1. Catalyst conditioning was first performed by operating the reformer at inlet temperatures of 610° C., 685° C., 735° C., 800° C., and 735° C., each for 8 hours. After conditioning, tests were performed on each of the catalysts at inlet temperatures of 685° C., 735° C. and 800° C.

[0058] Reformed gases were collected from the reformer and cooled to below the dew point to condense the steam and form condensates containing ammonia. The amount of ammonia in the condensates is proportional to the ammonia formed by the catalysts in the steam reformer. Condensate samples (250 ml) were collected over a period of 5 minutes at the end of the 8-hour test periods and analysed for their ammonia contents.

[0059] The ammonia concentrations in the condensates recovered from the reformed gases were measured using a calibrated Ion Selective Electrode (ISE). Standard solutions of 0.1, 1 and 10 ppm (w/v) ammonia were prepared. A sodium hydroxide buffer solution was added to the sample to liberate the ammonia. When the ISE voltage measurement was stable, the reading was used to generate a linear calibration curve of ISE voltage reading against log10 ammonia concentration. The ammonia concentrations of the condensates were analysed in the same way, using the ISE measured voltage reading to determine the ammonia concentration by derivation from the calibration curve.

[0060] The tests were repeated for each catalyst using feed gases containing different amounts of nitrogen. This was carried out by introducing nitrogen via a nitrogen supply line at various flows to provide the desired level in the feed gas fed to the reformer tube.

[0061] Tables showing the results of the ammonia produced in the condensates for the different catalysts for the different nitrogen contents in the feed gas are set out below.

TABLE-US-00002 Comparative Bed inlet [N.sub.2], [NH.sub.3], Catalyst 1(a) temperature ° C. vol. % mg/L 685 2 0.102 685 5 0.226 685 8 0.451 735 2 0.209 735 5 0.515 735 8 0.901 800 2 0.501 800 5 1.100 800 8 1.800

TABLE-US-00003 Comparative Bed inlet [N.sub.2], [NH.sub.3], Catalyst 1(b) temperature ° C. mol. % mg/L 685 2 0.050 685 5 0.081 685 8 0.144 735 2 0.071 735 5 0.210 735 8 0.420 800 2 0.166 800 5 0.435 800 8 0.807

TABLE-US-00004 Eggshell Bed inlet [N.sub.2], [NH.sub.3], catalyst 1(c) temperature ° C. mol. % mg/L 685 2 0.023 685 5 0.043 685 8 0.076 735 2 0.035 735 5 0.073 735 8 0.118 800 2 0.056 800 5 0.187 800 8 0.325

[0062] Over the range of inlet temperatures, Example 1(c) produces lower amounts of ammonia than the comparative examples. However, the catalysts contain differing amounts of nickel and have different activities. If a catalyst is more active, the amount of steam consumed will be greater than that for a less active catalyst. When this unreacted steam is condensed, it will affect the ammonia concentration. To account for this, a molar flow of water was calculated based on an oxygen balance derived from a knowledge of the feed gas composition and rate and gas-chromatography data on the reformed gas obtained using a GC system coupled to the steam reformer. The difference in the amount of oxygen entering and exiting the system can be used to determine the amount of ammonia produced per second.

[0063] Furthermore, each of the comparative examples contains more nickel. By measuring the amount of catalyst charged to the reactor precisely, it is possible to calculate the moles of ammonia per second per gram Ni.

[0064] FIGS. 1-3 depict molar ammonia concentrations of the condensates per second per gram Ni for the different nitrogen levels for each of the catalysts. This removes the differences caused by the different nickel contents of the catalysts and the varying amounts of condensate collected. It is evident that the comparative catalysts produce similar ammonia concentrations for each nitrogen level over the temperatures studied, but that the eggshell catalysts produced significantly lower amounts of ammonia, with the difference increasing in proportion to the nitrogen content of the feed gas.

[0065] The results for molar amounts of ammonia per second per gram Ni, for each of the nitrogen contents, at the inlet temperatures are set out below.

TABLE-US-00005 H.sub.2O NH.sub.3 make Inlet flow NH.sub.3 make, per g Ni, Comparative Temp., [N.sub.2], out, (×10.sup.−9) (×10.sup.−9) Catalyst 1(a) ° C. mol. % mol/hr mol/s mol/s/g 685 2 170 5.1 1.48 685 5 11.3 3.28 685 8 22.6 6.54 735 2 165 10.1 2.94 735 5 25.0 7.25 735 8 43.7 12.70 800 2 157 23.1 6.71 800 5 50.8 14.70 800 8 83.1 24.10

TABLE-US-00006 H.sub.2O NH.sub.3 make Inlet flow NH.sub.3 make, per g Ni, Comparative Temp., [N.sub.2], out, (×10.sup.−9) (×10.sup.−9) Catalyst 1(b) ° C. mol. % mol/hr mol/s mol/s/g 685 2 174 2.6 1.90 685 5 4.1 3.19 685 8 7.4 5.69 735 2 169 3.5 2.70 735 5 10.4 8.06 735 8 20.9 16.10 800 2 161 7.9 6.07 800 5 20.6 15.90 800 8 38.2 29.50

TABLE-US-00007 H.sub.2O NH.sub.3 make Inlet flow NH.sub.3 make, per g Ni, Eggshell Temp., [N.sub.2], out, (×10.sup.−9) (×10.sup.−9) catalyst 1(c) ° C. mol. % mol/hr mol/s mol/s/g 685 2 173 1.2 1.31 685 5 2.2 2.43 685 8 3.8 4.30 735 2 169 1.7 1.94 735 5 3.7 4.09 735 8 5.9 6.58 800 2 161 2.7 2.98 800 5 8.9 9.94 800 8 15.4 17.3

[0066] The temperatures depicted in FIGS. 1-3 have also been adjusted to reflect the average bed temperatures, which were calculated by taking an average of thermocouple measurements taken at the same height as the catalyst bed inlet and catalyst bed exit. This takes into account the larger endotherm observed for the more active catalysts. The results illustrate that, even taking the different nickel contents and the endotherm generated by the reforming reactions into account, the eggshell catalyst 1(c) out-performs the comparative catalysts 1(a) and 1(b) in terms of ammonia produced. Moreover, the eggshell catalyst 1(c) was able to produce a reformed gas with a high conversion of the hydrocarbons in the natural gas.

[0067] The reformed gas after condensate removal was analysed by gas chromatography to establish the conversion of hydrocarbons to hydrogen and carbon oxides. The conversion of the ethane in the natural gas is a better measurement of overall catalyst activity than methane conversion, which is reversible. The results are set out in the following Tables:

TABLE-US-00008 Catalyst Comparative Catalyst 1(a) 2% N.sub.2 5% N.sub.2 8% N.sub.2 Ethane [NH.sub.3]/ [NH.sub.3]/ [NH.sub.3]/ Inlet conversion (×10.sup.−9) (×10.sup.−9) (×10.sup.−9) Temperature (%) mols.sup.−1 mols.sup.−1 mols.sup.−1 685° C. 52.19 5.1 11.3 22.6 735° C. 59.50 10.1 25.0 43.7 800° C. 74.12 23.1 50.8 83.1

TABLE-US-00009 Catalyst Comparative Catalyst 1(b) 2% N.sub.2 5% N.sub.2 8% N.sub.2 Ethane [NH.sub.3]/ [NH.sub.3]/ [NH.sub.3]/ Inlet conversion (×10.sup.−9) (×10.sup.−9) (×10.sup.−9) Temperature (%) mols.sup.−1 mols.sup.−1 mols.sup.−1 685° C. 32.71 2.6 4.1 7.4 735° C. 39.64 3.5 10.4 20.8 800° C. 57.14 7.9 20.6 38.2

TABLE-US-00010 Catalyst Eggshell Catalyst 1(c) 2% N.sub.2 5% N.sub.2 8% N.sub.2 Ethane [NH.sub.3]/ [NH.sub.3]/ [NH.sub.3]/ Inlet conversion (×10.sup.−9) (×10.sup.−9) (×10.sup.−9) Temperature (%) mols.sup.−1 mols.sup.−1 mols.sup.−1 685° C. 35.50 1.2 2.2 3.8 735° C. 44.20 1.7 3.7 5.9 800° C. 57.25 2.7 8.9 15.4

[0068] These results illustrate that the eggshell catalyst 1(c), despite having a lower nickel content than Comparative Catalyst 1(b), is more active for the reforming reactions. Comparative Catalyst 1(a), which has more than three times as much nickel, is more active, however the ammonia produced is significantly higher at 2%, 5% and 8% N.sub.2. This is illustrated in FIGS. 4, 5 and 6 by plotting the condensate ammonia concentrations obtained against the ethane conversion.