NL8303596A - METHOD FOR REMOVING METAL POLLUTANTS FROM HYDROCONVECTION CATALYSTS - Google Patents

Description

VO 5178

Method of removing metal contaminants from hydroconversion catalysts.

The invention relates to the removal of metal impurities from hydroconversion catalysts. More in particular, metal contaminants are selectively removed from catalysts on refractory oxide supports by treatment with buffered oxalic acid solutions just before the support is dissolved.

Hydroconversion catalysts for the treatment of petroleum feedstock are inactivated by such factors as carbon formation and metal impurities usually present in crude oil. It is known to use acids to regenerate the hydroconversion catalysts. U.S. Patent 2,280,731 relates to the removal of iron and other metals from catalytic cracking catalysts by treating the spent catalyst with an organic acid and a dilute mineral acid. Preferred are aqueous solutions of oxalic acid. U.S. Patent 3,020,239 describes the removal of vanadium from molybdenum-containing catalysts using an aqueous glycolic acid solution. Other hydroxy acids and compounds analogous to glycolic acid are unsatisfactory because of the simultaneous removal of molybdenum. It is preferable to adjust the concentration to prevent the leaching of aluminum from the support material. US Pat. No. 3,791,989 teaches to remove vanadium from a hydrotreating catalyst containing a Group VI or Group VIII metal by contacting the catalyst with an aqueous solution of oxalic acid before burning the coals. In Example 1, precipitated catalyst particles were treated with a boiling concentrated oxalic acid solution removing preferential vanadium over nickel or molybdenum. The reverse order is described in British Patent 1,245,358 in which a deactivated supported Group VI or VIII catalyst is first subjected to carbon combustion and then washed with an aqueous oxalic acid solution at a concentration of 0/5 M to saturation point. When the contact time with oxalic acid is limited, no catalytic metal is produced.

3303596 «» 2. removed.

In U.S. Pat. Nos. 4,089,806 and 4,122,000, hydrodesulfurization catalysts containing Group VIB and / or VIII metals on a refractory support are regenerated with a combination of oxalic acid plus nitric acid and / or nitrate salts. When only oxalic acid is used under conditions sufficient to remove substantial amounts of vanadium, it has been found to lead to the dissolution of the alumina carrier. Finally, according to U.S. Pat. No. 3,536,637, iron-exchanged ion-exchange materials are regenerated by contacting the resin with oxalic acid.

While it is thus known that oxalic acid is suitable for the removal of certain metal contaminants from hydroconversion catalysts based on inorganic oxides, such as alumina, oxalic acid has the drawback of also removing or dissolving catalytic metals and support materials.

It has now been found that the above-mentioned disadvantages can be overcome without this having any detrimental effect on the ability to remove oxalic acid metal impurities when buffering the aqueous oxalic acid solution. Thus, the method of the invention for removing metal impurities from a hydroconversion catalyst containing at least one Group VIB, VUB VIII metal supported on a refractory inorganic oxide includes contacting the contaminated catalyst with a fumed aqueous oxalic acid solution. In another embodiment, the metal impurities are removed from a catalytic cracking catalyst, which cracking catalyst contains at least alumina, silica, silica, silica-alumina, zeolites or clays, according to a method in which the contaminated cracking catalyst is contacted with a buffered aqueous oxalic acid solution .

In the present method using a buffering agent, it is possible to use oxalic acid at different concentrations without dissolving the metal oxide supports, as commonly used in hydroconversion catalysts.

In addition, long contact times of the oxalic acid solution with the contaminated catalyst are possible without removing catalytically active metal 8303596 4% 3 from the support, even at elevated extraction temperatures. With buffered oxalic acid, it is also possible to regenerate catalytic cracking catalysts without dissolving the catalyst.

Hydroconversion catalysts of interest are those in which either the active catalytic component or the support material is susceptible to attack by inorganic and / or organic acids. In normal commercial use, metal and carbon deposits form on the catalyst, which inactivates it. The inactive catalysts are regenerated with an acid to remove the metal impurities. However, if the support or catalyst is attacked by acid, difficult problems arise because loss of support can also lead to loss of active catalyst component. Such losses are often unacceptable or the contact time between acid and contaminated catalyst is so limited that sufficient amounts of metal contamination cannot be removed.

Hydroconversion catalysts of particular interest include cracking catalysts, reforming catalysts, and residue conversion catalysts. Catalytic cracking catalysts are Sure metal oxides such as alumina, silica, silica-alumina, zeolites, clays, especially acid-treated clays, silica-zirconia, silica-magnesia, alumina-boria and silica-titania. The combination of the above cracking catalysts as carrier with a hydrogenation dehydrogenation metal, for example, Co, Ni, W, V, Mo, Pt, Pd or combinations thereof, gives hydrocracking catalysts. Generally, the combination of a Group VIB metal, Group VUB, Group VIII metal or mixture thereof supported on an acidic support as described above is used for various catalytic hydroconversions, such as residue conversion and reforming. The groups are defined in the Periodic Table on page 662 of the 9th edition of the Condensed Chemical Dictionary. Preferred catalytic materials are alumina, silica-alumina or zeolites as carriers Co, Mo, Re or platinum group metals, either alone or in combination.

Metal contaminants come from either the feed material or the metal components used in the manufacture of reactors and transport lines. Examples of metal impurities which can be removed by oxalic acid are 8303598-4 * 4 iron, vanadium, nickel, sodium, magnesium, chromium, copper, strontium, lithium, lead and the like. Such metals are usually undesirable since they can lead to catalyst inactivation by catalyst poisoning, occlusion or both.

As noted above, an oxalic acid extraction to remove metal impurities can also lead to acid attack of susceptible carriers, active catalytic metals or both. In addition, some oxalic acid treatments require high temperatures and / or acid concentrations that aggravate the above-mentioned problems. It has now been discovered that the above problems can be avoided by buffering the oxalic acid without affecting the ability of oxalic acid to extract the metal impurities.

The buffering agents of the invention are capable of buffering 15 aqueous oxalic acid solutions within a pH range of 2 to 10, preferably 4.5 to 8.5. It is preferred that the buffering agents do not contain Group IA or Group IIA cation in amounts such that the catalyst is poisoned.

Examples of preferred buffers are ammonium salts of weak acids, for example ammonium acetate, ammonium citrate, ammonium formate, ammonium lactate, ammonium valerate, ammonium propionate, 4 ammonium urate, etc., or mixtures thereof. Mixtures of different buffers are known to extend the pH range to be adjusted and to compare with the individual buffers. The use of ammonium salts of weak acids prevents the possible adverse effects of adsorption of the cations of bases and alkaline buffers. It is known that such cations are strong poisons for many hydroconversion catalysts.

The use of metal chelating buffers is also possible to promote the selective removal of metal impurities from hydroconversion catalysts. Oxalic acid alone is known to form chelates with niobium and zirconium and remove them from ion exchange columns ("The Chemistry of Lanthanides", T. Moeller, Reinhold Publishing Corporation, 35 New York, 1963, p. 81-85). Buffers that have different metal binding capacities are: NH2C0CH2NHCH2CH2S03H, NH2COCH2N (CH2COOH) 2,8303596 Bicine, Glycylglycine, O NC ^ C ^ SQ ^ H, Tricine. ("Buffers for pH and Metal Ion Control /" D.D. Perrin and B. Dempsey, John Wiley and Sons,

New York, biz. 108). Glicin derivatives are known to be strong complexing agents relative to iron and other metals. It is also known that hexanoic acid and di (o-hydroxyphenyl) acetic acid are strong complexing agents for iron and other metals and can be added to buffer compositions ("Photometric and Fluorometric Methods of Analysis. Metals Part 1," FD Snell , John Wiley and Sons,

New York, 1978, p. 763). By using a strong complexing agent in the buffer solution which improves the rate and extent of the metal contamination removal from a hydroconversion catalyst, in addition to the oxalate ions, an improved mefcal contamination extraction procedure can result.

Another group of buffers for the metal contamination removal from hydroconversion catalysts are buffers useful in aqueous / non-aqueous solvent systems. For example, in 90% MeOH / 10% 10 O solution, a 0.01 molar oxalic acid, 3 0.01 molar ammonium oxalate will buffer the pH at 4.23. In a 60% MeOH / 40% solution, this same buffer has a pH of 2.58. (See "Buf-20 fers for pH and Metal Ion Control," D.D. Perrin and B. Dempsey, John Wiley and Sons, New York, 1974, pp. 84-88). Aqueous / non-aqueous systems are useful for removing metal impurities from hydroconversion catalysts since alumina solubilization is greatly impeded compared to aqueous systems.

In some cases, the buffers may contain small concentrations of group IA component. In such a case, these buffer systems can be useful for removing metal impurities from hydroconversion catalysts. Examples of such aqueous buffers are phenylacetic acid and sodium phenylacetate, acetic acid-30 and sodium acetate, succinic acid and sodium hydroxide, potassium hydrogen phthalate and sodium hydroxide, sodium hydrogen maleate and sodium hydroxide, potassium dihydrophosphate and sodium hydroxide, sodium hydroxide and water hydroxide and water hydroxide. All of the aforementioned buffers have a pH range that is recorded by the buffer component concentrations.

Other aqueous buffers useful in conjunction with oxalic acid to remove metal impurities from hydroconversion catalysts include such aqueous buffers as glycine and hydrogen chloride, imidazole and hydrogen chloride, 2,4,6-trimethylpyridine and hydrogen chloride, N-ethyl morpholine and hydrogen chloride, tris (hydroxymethyl) aminomethane, 2-amino-2-methylpropane-1,3-diol 5 and hydrogen chloride, diethanolamine and hydrogen chloride, as well as ammonia and ammonium chloride. In some cases, a wide range buffer (pH 2.6-8.0) such as citric acid and N-a2HPO4 may also be useful for removing metal impurities from hydroconversion catalysts. Other well-known wide range buffers are: piperazine dihydrochloride, glycylglycine and sodium hydroxide; citric acid, K2HPO4, diethyl barbituric acid and sodium hydroxide, boric acid, citric acid and tri sodium orthophosphate.

The amount of buffer required depends on the concentration of the oxalic acid used. Oxalic acid concentrations can range from Ο.ΟΟΟΙΜ to 5.0M, with concentration ranges from 0.01 to 2.0M being preferred. The amount of buffer should be sufficient to maintain the pH of oxalic acid within the desired range. The increase in the concentration of oxalic acid will generally require a large amount of "a certain buffer, depending on the buffer capacity * 20".

The temperature ranges from about 0 ° C to about 100 ° C, and is preferably 20-75 ° C. Higher temperatures can lead to excessive acid attack of the support and / or the catalytic metal without offering any particular advantage as regards the removal of metal contaminants.

Unlike conventional oxalic acid, buffered oxalic acid can be contacted with inactivated catalysts for long periods of time without dissolution of support materials. The method of contact of the catalyst and the buffered oxalic acid solution is not critical. The inactivated catalyst can be soaked in a buffered oxalic acid solution. Samples of catalyst or buffered oxalic acid solution can then be removed at fixed intervals to monitor the degree of metal contamination removal.

Optionally, the catalyst can be extracted continuously, such as in a countercurrent extractor with buffered oxalic acid recirculation.

The preferred embodiment for removing metal impurities from hydroconversion catalysts involves the regeneration of refonn catalysts typically containing Pt, either alone or in combination with another metal, e.g. Re or a noble metal, on an acidic support such as alumina or silica-alumina.

The refonning catalysts are gradually contaminated with metals such as Fe. Pb, Cu, Ca, Na and the like, in which Fe is particularly troubling.

When 0.5 M oxalic acid is suspended with Fe-contaminated Pt, Pt / Ir or Pt / Re catalysts on γ-Α ^ Ο ^ and heated, .a significant amounts of carrier and precious metal are combined with the desired Fe impurity deleted. However, the ammonium acetate buffered oxalic acid extraction from the same catalyst systems results in selective Fe removal, with only very small catalyst losses due to mechanical losses rather than chemical attack of the γ-alumina support.

Lower temperatures of 20 to 75 ° C are preferred when Fe is removed using buffered oxalic acid solutions.

After the extraction, a Pt / Re catalyst was calcined in the presence of at 500 ° C for 4 hours. The regeneration catalyst and CO chemistry values were almost the same as that of the fresh catalyst, indicating that no poisoning or surface modification occurred due to the buffered oxalic acid treatment.

A Fe-extracted Pt / Ir catalyst was compared to the non-extracted catalyst containing 0.35 wt% Fe and a fresh Pt / Ir catalyst for the naphtha reforming activity. After 400 hours, the extracted catalyst showed only a 25% low reforming activity with respect to the fresh catalyst compared to a 70% decrease in reforming activity for the catalyst that had not been extracted with a buffered oxalic acid solution.

The method of the invention is further illustrated by the following examples.

Catalysts

The catalysts were supported on γ-Al 2 O 2 supports with a BET specific surface area of 150-190 m2 / g.

35 Pt / Re / Al ^ O ^. The platinum-rhenium bimetallic catalyst was a commercial sample with an initial composition of 0.3% Pt, 0.3% Re and 0.9% Cl (all percentages are by weight unless indicated otherwise 8303598 8). ).

Pt / Ir / AJUCL ,. The platinum iridium bimetallic catalysts used were commercial samples. The fresh catalyst contains • 0.3% Pt, 0.3% Ir and 0.7% Cl. A number of used Pt / Ir / 5α12 ° 3 samples were also obtained from commercial refineries.

Pt / Αΐ ^, Ο- ,. The monometallic platinum catalyst was a commercial sample with the composition of 0.3% Pt and 0.7% Cl.

Iron-doped catalysts. Known amounts of Fe were added to fresh Pt / Re / Al 2 O 3 and Pt / Ir / A 2 O-3 catalysts until they started to get wet using standardized aqueous iron nitrate solutions. After drying in air at 120 ° C for 16 hours, the impregnates were calcined at 270 ° C for 4 hours under 20% C 2 / He (500 cc / min) to ensure decomposition of the nitrate salt.

Example I

This example demonstrates the deleterious effect of Fe on the chemisorption properties of. precious metal bimetallic catalysts. Hydrogen and carbon monoxide chemisorption studies were conducted with a conventional glass vacuum system as described by Sinfeit and Yates in J. Catal., 8, 82 (1967)

The hydrogen and carbon monoxide recordings were determined at 25 ± 2 ° C on the reduced and evacuated samples. Typically, 30 minutes were allowed for each recording point. The H / M ratios were calculated by assuming that the hydrogen uptake at a zero-25 pressure corresponds to the saturation coverage of metal. Hydrogen uptake at zero pressure was determined by extrapolation from the linear high pressure portion of the isotherm as described by Benson and Boudart and Wilson and Hall (J. Catal., 4, 704 (1965) and 17, 190 (1970) The CO / M ratios were calculated by determining the carbon monoxide recordings on the reduced and evacuated samples and assuming that this amount is the sum of weak A ^ 2 ° 3 <rar and star ^ represented carbon monoxide bound to the metal surface The sample was then evacuated (10. Torr) for 10 minutes at room temperature and a second carbon monoxide isotherm was measured Since the second isotherm 35 measures only the carbon monoxide weakly adsorbed on the support, a subtraction of the two isoterms the amount of carbon monoxide which is 8303596 W9 strongly associated with the metal components The amount of strongly bound carbon monoxide at 100 Torr was chosen as the saturation coating of the metal.

The hydrogen chemisorption properties of fresh and iron-doped Pt / Re and Pt / Ir catalysts are summarized in Table A.

TABLE A

hydrogen chemisorption properties of iron-doped Pt / Re and Pt / Ir __ alumina catalysts_

Catalyst% Fe (a) H / M (b) D / Doic) 10 0.3% Pt / 0.3% Re / Al203 0 0.4 100 0.10 0.3 75 0.36 0.2 50 0 .3% Pt / 0.3% Ir / Al2 O3 0 1.7 100 0.05 1.4 82 15 0.10 1.3 76 0.36 1.0 59 0.80 0.8 47 a) Iron added like the nitrate salt. After the iron impregnation, the catalysts were calcined at 270 ° C under 20% O 2 / He (500 cc / min) for 20 4.0 hours to ensure decomposition of the nitrate salt.

b) Atoms of hydrogen chemically sorbed per metal atom in the catalysts at room temperature. Before the chemisorption measurements, the catalysts were reduced at 500 ° C under H 2 (500 cc / min) for 2.0 hours.

c) Normalized dispersions

In the last column of Table A, the hydrogen uptake of the iron-doped catalysts has been normalized with respect to the uptake of the fresh Pt / Re and Pt / Ir catalysts. For both bimetallic catalysts, systemic reductions in hydrogen chemisorption capacity were observed with increasing iron concentrations. At an iron concentration of 0.36%, the hydrogen uptake of the bimetallic catalysts was 40 - 50% lower than that of the fresh 8303596

«V

ίο monsters.

The Fe / noble metal mole ratios for catalysts containing 0.1 - 0.8% Fe range from 0.5 to 4. Thus, a 0.2% iron concentration would be high enough not to interact 5 all precious metal components present in Pt / Re and Pt / Ir reforming catalysts. In fact, the significant reductions in H / M values caused by the low iron concentrations indicate that iron either alloys with or physically blocks a significant fraction of the noble metals. A decreased chemisorption capacity can potentially reduce reforming activity since H-H and C-H bond activation processes are extremely important in catalytic reforming.

Example II

The effect of iron on the reforming activities of bimetalic catalysts is shown in the example. Naphtha reform reactions were conducted in a 25 cc, stainless steel, fixed bed, isothermal hydrotreating unit operating in a single pass.

The reforming tests were performed at 487 - 489 ° C under a total pressure of 14 kg / cm2. The hourly space velocity in weight 20 was 2.1 WHW and hydrogen was fed in an amount of 1068 SCF H 2 / vol. A commercial naphtha feed was used in which the sulfur concentration was adjusted to 0.5 ppm by adding thiophene. After the hydrogen reduction at 400 ° C, the catalysts were sulfided using a dilute S / H mixture. The reformate 25 was analyzed for research octane number (RON). The octane numbers were used to define relative catalyst activities (RCA).

Naphtha reform activities of fresh and iron-doped Pt / Re and Pt / Ir catalysts are compared in Table B.

30 8303596 11 V *

TABLE B

Naphtha reforming activities of iron-doped Pt / Re and Pt / Ir on alumina _________ catalysts__

Catalyst (a)% Fe Hours in RON (b) RCA (c) ______ _ operation _ _ 5 0.3% Pt / 0.3% Re / Al203 0 158 99.3 81 278 99.1 74 422 97.6 56 0.3% Pt / 0.3% Re / Al203 0.36 162 95.3 47 282 90.6 33 10 0.3% Pt / 0.3% Ir / Al203 0 165 101.5 128 285 101, 6 125 309 101.3 126 0.3% Pt / 0.3% Ir / Al203 0.36 166 97.0 58 285 93.4 46

(A) Same catalysts as described in Table A.

(b) RON 3 research octane number (c) RCA = relative catalyst activity

The relative catalyst activities clearly demonstrate that 0.36% iron has a marked adverse effect on both bimetallic catalysts. Similar carbon deposits were found for all catalysts. Due to the 55-65% lower activity of the iron-containing catalysts, it is shown that the precious metals are inactivated by interaction with iron. Since in these catalysts the Pe / noble metal to molar ratio is close to 2, sufficient iron is available to modify all the Pt / Ir and Re. The reduced reforming activities are consistent with the reduced hydrogen chemisorption capacities found for the iron-doped catalysts.

Example III

3D The solubilities of γ-Αΐ203 in H 2 O, oxalic acid and ammonium acetate buffered oxalic acid solutions are compared in Table C.

8303596 12

TABLE C

Solubility of γ-Α ^ Ο ^ in aqueous oxalic acid and buffered oxalic acid solutions

Treatment ^ pH Solution__ Recovered γ-Αΐ ^, Ο ^ (%) ^ T ° C million ^ 0 O.A. BOA. H ^ O O.A. BOA.

5 100 15 7.6 0.6 4.2 100 85 99 100 30 4.9 2.0 4.4 100 52 100 (a) 5> 0 g of γ-Αΐ ^ Ο ^ under reflux in 50 ml aqueous solutions of net HjO, 0.5M oxalic acid (OA) and 0.5M oxalic acid plus 0.5M ammonium acetate (BOA). γ-AljO2 in the form of 0.16 cm extrudates.

(B) The recovered catalysts were rinsed with four 50 ml parts of distilled HjO and then dried at 120 ° C for 16 hours.

After the reflux treatment in H 2 O for 30 minutes, the pH of a γ-A 2 O 2 suspension had reached a self-buffer level of about 5.0. The acidic nature of A 2 O 3 -Q suspensions is well known in the art. A suspension of AljO4. oxalic acid shows a pH of about 2 at 100 ° C. Thus, the acidity of an Al 2 O 3 oxalic acid suspension is approximately three orders of magnitude greater than that of an Al 2 O 2 O 2 suspension. It was observed that an Al 2 O 3 ammonium acetate buffered oxalic acid suspension, in contrast to the oxalic acid suspension, maintained relatively constant pH values of about 4.4. Thus, the pH of the buffered suspension is of equal value to that of an aqueous suspension of A ^ O ^. The effect of the pH is reflected in the amounts of A ^ O ^ that can be recovered unchanged from the suspensions. An essentially complete recovery of A 2 O 2 was achieved from the 2 2 and buffered oxalic acid suspensions. However, oxalic acid suspensions easily dissolve significant amounts of A 2 O 3. After a reflux treatment for 30 minutes, almost half of the initial amount of A 2 O 2 had been converted to soluble aluminum compound.

Example IV

This example illustrates the non-selective extraction of iron from precious metal catalysts. The results of representative studies using oxalic acid solutions for extracting iron from Pt, Pt / Ir and Pt / Re catalysts are summarized in Table D.

8303596 13 ο o

• P

YOU

Ό 0 - c ο Φ c co o cm m ω S c othOOO '^ o 3 m o a) ·. v i-i 0 U Hi Ο O OO Ο- * Ο Φ 4J 0 X,

s is I.

SR „* 5 d o '« I I I ii i i ° o 1 1 “SI I I. 1 1 1 o i «r * 3 .c g Ή id fi ® ®, 3 ®

ij 0 IQ 1-1 Π3 S

φ O 'μ id ο »ν · Γ' in d ® g β Φ +> l I * - <(min dj • Η Φ Φ H ii '' '' I * d ® m, cg oo-oo ohh oi d - g • δ H +1 · Λ ® 0 ο · φ φ .or- mo in ιο m ο Φ φ Μ ® O '+> l CM * <«Ί H CM' -l CM O 'irt 3 Q v Oil * * »- -" β - C 3 1 OO OO OO od β 3 5> ggs § β SS £ 5 J -8 ° Λ -5 Φ Η β φ ® ^ 3 Ο θ 'β g, 3 d μ β Μ ^ η δ> 3 3 Φ β Ο d? Οι ^ .5 + 1 ojj .ON.CIOJ OO £ ο Ο ^ ε d 3d 5 cn r- t ^ tn σ '("ω m.2 3 a φ to φ ® J ^

Μ Ο »> O '5 m IU

* g § · * 3 ί ί η Ü> ®dss> § => 5 S 2 * ί ο *, 5 4J φ Ρ 0) m Η XS Η 2 ® 8 Φ Ό Μ Φ β Φ ^ Λ ~ β ο ο οο οο Ο § ο τ) S 2 * ω -2 * * * C®? > J - (Μ _ yy ο s Η <*> ο ® s.

Jj rrt * Η Φ * ΐί 3 λ β μ τι s ο § μ d ο η ® § Λ Ο Ο Ο ΟΟ Ο Ο Ο φ> +) Λ ο μ φοιηο οο «η Ο«> - g - cm jj m ii -ί -Ί <- «t — t bwmv} Z wp MM> ι Φ o S 0 0 -) ¾ <“ + J 5 d β omd +) h S mof * oj φ d vo S - φΐη ω ο n > Ό ^ 3 XJ 3 ^ 1 (3 o cm odcfidr *

η Λ 4J -Η Φ -a N

AJ ll r <d Cj Ö 3 !! obsess <? m5S | i! i ^ 01 o ® 5 e Λ>, φα | Ο Ο μ O 'Φ Φ

g j? 2 - d O 'O' M

5 e ”m tn 3 O 'S Jj M>, M d β

Zd · σ »ρ * I r-j a> c ω

1 «1 1 CM CM ioYy ^ dO

m © η i - »I ° g 2 3

O 'O' ο o '£ J3 P

C w n m a Ό fe _ Οϊ G \ & GO Ο »

μ, CM CM CM CM - - -O

• Η μ - - '' d Λ 3

Dftlo o o O ~ 8303596

V

14

After 60 minutes at 50 ° C, approximately 10% of the support is solubilized by a 0.5M oxalic acid solution. By extractions performed at 100 ° C, 30-40% of the A 2 O 3 support is dissolved in 60 min. These significant carrier losses are too high to seriously consider oxalic acid as an extractant for metal impurities. During the extractions performed at higher temperatures (about 100 ° C) a measurable fraction of the noble metal components were extracted together with the iron. The loss of Re is particularly noticeable and cannot be adjusted by lowering the extraction temperature or reducing the concentration of the oxalic acid solution. Example V

In contrast to Example IV, a selective extraction of iron from precious metal catalysts using buffered oxalic acid is indicated herein. The results of studies 15 ammonium acetate buffered oxalic acid solutions for extracting iron from Pt, Pt / Ir and Pt / Re catalysts are summarized in Table E. Before extraction with buffered oxalic acid solutions, the iron-contaminated catalysts were typically reduced under H 2.

8303596 I 15 -r * «3 • μ C ·« ο

P C

ë <β μ? (0 <β ρη οι ν Ο C0 -Ρ <0 S 'o ^^. IncnvocMO a Ό • η e φ I - «-» λ "* * ym CO ® En | Ο Ο O τ-iOt-i ^ ogc η μ § w 5 R -ri o 8 * e O 'com tul III lilies ® 2>, SI II 1 1 1 1 1 ο 1 Ό μ λ 5 10 ο a £ ~ S®' SS Λ ffl Λ £ · * 'P hu φ © co oo σι ω o -h

c> 3 h p | ι l M fOCniNCM j -P

2 SJS "I 1 1 o o o o o 'j I

I 18 * ï (0 <0 I, μ m * g ö s. 2 5 j? X> cs o o o o ^ co o d jö (rf ίφ o ^ j | n n m cn cm cn cm η N> ·

g Sp o o ooooo o «e S

3 S ΐ |

3 »a I

axis 0 S

® c o | ^ 5 ^ Ή 0 a «we -π p η φ φ S 'J" d * 3 ep - cS' ^ ® § §5 * 0® to co r * * co <£ -H -P -¾ 2 ê <® ïï ί οι θ 'σ' σι σι o σι σι 2 $ s <ö 4J φ tn S ^ Ώ 8 ^ 2 '& 3 3 p «a

2 S 'd ft +> H -H

P -P1 o x a

Ώ S tJi Φ 0 C

5 £ S φ a ® o ** 1 co a CO Ό · 1-i O Φ p

e POs

e φ O S

β Λ -P «h C

§ O 'ü 3 30¾¾ ^ * 1 SJ 5 S § S S WS S1 § dg S 8 | +) co O '8 8 S 3 £ 8 • dOC o> ca 3 ft 3 ^ 8 2 01 J.

«§ f 'you'? 5 E ° a ° *

N3C i «ü (öpeacM

Ï? N λοοο o oooo o -P a «c '-o

i-4 φ o tn o tn minoo tn "d ^ i? iq coO

Örn Cj, H * 4 M r-i \ £ J Ui W

2 w PC (Ö * - · P

g S φ -P -P Φ -H

Cn ° C Λ Ό OT C

S m m cn r'PTi C C - <7 pp Λ S | ^ ^. ^ 2 N | § δ «

> § <poS ° One ° gCC'OaC

J 3 «« already i 'I! ^ poa Sd g -g 2, s “1” 1 o SB ft o 5 «S ft * 5« s §> s, g ft> Jj Jj Jj _> 1 P <> 1 14 c ΦΦΦΦ, vor-, S3 «SS! "" ioS «3s!

J J SC ° ° σϊ S S a 2 S

Φ | §I Φ σι coco σι rHC + JCnCM CNCS ™ 'Z' n "o« ί i 31 S o- o o o S e - 2 8303596 16 *

Compared to oxalic acid suspensions, buffered oxalic acid solutions did not solubilize the A 2 O 3 support.

The apparent catalyst losses of 2-4% are mainly due to mechanical losses from filtration of the slurry instead of the solubilization of A 2 O 2 in the buffered solution. Iron is selectively extracted and no precious metal removal takes place, even at extraction temperatures of 100 ° C. The extraction data further indicate that iron removal is promoted at a lower treatment temperature. This behavior is consistent with the fact that the iron oxalate complexes are more stable at lower temperatures.

The effect of the catalyst particle size on the extraction efficiency is shown in the case of a 2% iron-doped Pt / 1r catalyst. Smaller catalyst particle sizes can be expected to expose more surface area to the extraction media. Thus, under a certain range of extraction conditions, the removal of iron and other extractable impurities at smaller particle sizes becomes more favorable. The improved removal of iron from the powdered catalyst is in accordance with these arguments.

The extraction data summarized in Table E was obtained from crude slurry tests, which were "not optimized. However, it is reasonable to assume that essentially all of the iron contained in a particular catalyst can be removed by either transferring the catalyst to the fresh to recycle extraction media or by designing an extraction procedure whereby the catalyst is continuously contacted with a buffered oxalic acid solution The use of buffered oxalic acid solution can also be utilized to remove iron and other extractable metals such as Na, 30 V, Ni, Cu and others from a wide variety of hydrocarbon conversion catalysts Although oxalic acid removal is preferred for the removal of metal impurities, it is possible to include other organic acids in the process of the invention.

Example VI

It is known that iron can act as a poison for precious metal catalysts. This example demonstrates the beneficial effect of the 8303596 17 C iron removal on the chemisorption behavior of a Pt / Re catalyst.

The results are shown in Table F. 1 8303598 «« 18

I I

o tn I

c SO M

Φ g Η M 3 'Z'

M O (d O4 O + J 3 t-> C

- 0 H O 44 Φ N Η H

44> 0 44 M UlgH OS

0 -H 0 3 C _0

OS 3 STS SR

>, S N S MX 0 0

£ O <0 H BOO M

0 g μ o 3 o q ο

yy g φ Ο H A 0 44 O

(0 Ο Ο X Μ Ο Ό (din 44 MO Β Ο Μ Ο Μ M M w 1 φ id Ο P P O> i

H 44 Φ MO) 4) ο X <w H CN

(d 1) Ό -ΡΌ cn M 10 Ή (dB

O · Φ β Μ # Μ C Ο Ο 3 -3 4J Η Ο Ο 100 Ο Λ θ 'Λ id Μ

Ο) MgOOM-lSidOU ^ O

δ Ο 5 »Ο» Ή ι-4 -. Ο »ο Ό

Ρ Η Μ 3 Η _ Η Ο Ο G

Ο) Φ φ φ -> - 43 Ο Ό ^ 44 m Ό Ο η η φ ft no om id

φ p Η 01 - ’Ο» ϋ -Η Ο fi (d Tl CU

η £ 1ϋ-> 4 ο Ο Ο 1) 4J 4 Η ° Ο ΜΜΌΜΜ44ΟΜ4-) Μ0Λ “2 ooooo <dmoQOo_ so Ο 4-) 6044440 g'öpfdö 'om

Ό (dS (d Id1) n S O 0 S G O

C oiOStnMO'HOO'tnOiH 44 -n

3> iü3> i> iOPOI> i-Hcn O -W

0 Η φ H HO Μ H C 01 H -Q

p O 01 HO 080 WO OOO o

01 4JM04444CCMN44SH OC

Μ O Φ ft O O Η H φ fl O S ft 44 φ 0 m> O4444fi01> ι44400 0¾ 1 B $

0 MO

1 s a O ft ft C - Γ1 H «-1 Ό 0 SO M 44 § w φ o Β H C O 44

O O P

ft G »^« j - £ O Ό CQ φ i4 tn

Cm Ό S cm tn r »r- m m OC

4 ^ 3 \ 1 1 1 · 1 · 1 O 'O

000 ^ 0 0 O O O Ό

E-ι φ 43 U. Ρ M

• PM O O

O O 'M 3

Μ> P

44 I SC

* i-4 O O

ιφ O Λ O '

φ O O C

O '44 1H

φ O 44 MS V <8

φ S> t S

N 44 1 I

• n O O O

H O S> -H CO 1 CN f β H

P '-v. > 1 1 · «1« · · 1 OP

C Μ Β η h O OOO Sft

0 0 i-4 M

> 01 0 0

φ O CO

g 1 1 ï o! o Ρ τ5

ff HOM

p 3 0 o m o p'co ooo and o im h i m o o Ό o C fci I | · »1 i · 1» 1 »iH 3 o c o o oo 2.¾13 O 'o S jo o

Η Η Ό M

φ «j o 'crv c> o | 44 m CN CN Φ O 'φ φ φ I i i i - - - · H g «| 1 I I OOO "yü3

44 q 3 O

ft dP 44 3M

u tn 44 3

o 3 10 CO MO

(OOMlmCN CM lil O M CN

14 O 'H 1 1 »III 44 O

6 _ o o o oao Φ C 3 S Ό

Po o c OM CO o O'O C 44 φ O 441 m 04 m mcNcn ΦΜΜ 44 A | 1 1 »'·' · 'S O 3 O OO o OOO O S Ό UI p o o> 1 C <1 O' r14 10 Λ fö <PQÜ Q S i 'ti w w w WWW w 8303596 19

The addition of 0.37% iron to a fresh Pt / Re catalyst reduced Hj and CO uptake by approximately 50%. Extraction of the iron-doped catalyst with an ammonium acetate buffered oxalic acid solution at 50 ° C decreased the iron concentration from 0.37 to 0.08%. The extracted catalyst was run at 500 ° C

Calcined under 20% C 2 / He for 4 hours to ensure the removal of the last traces of extractant from the catalyst surface. After the extraction-calcination steps, the chemisorption uptakes had increased to values close to that of the fresh catalyst.

The restoration of the chemisorption values of the Pt / Re catalyst upon removal of iron indicates that the buffered oxalic acid extractant does not poison or modify the metal surface.

The redispersed Pt / Ir catalyst showed hydrogen and carbon monoxide uptake lower than that of a fresh catalyst. X-ray diffraction measurements on the redispersed catalyst did not reveal any Pt or Ir diffraction patterns, so it is reasonable to assume that the catalyst is well dispersed. Thus, the presence of 0.35% iron on the redispersed catalyst may be responsible for the abnormally low hydrogen and carbon monoxide uptake. Extraction of the iron from the catalyst with an ammonium acetate buffered oxalic acid solution lowered the iron concentration from 0.35 to 0.10%. After the standard extraction procedure, the catalyst was calcined at 270 ° C under 20% (1 / He) for 4 hours.

The iron extracted and calcined Pt / Ir catalyst showed hydrogen and carbon monoxide uptake slightly lower than that of the redispersed catalyst. Since the iron-extracted catalyst was only subjected to a 270 ° C calcination step, the low chemisorption values may arise from residual amounts of the extractants which adhere to the metal surface.

However, the low chemisorption values exhibited by the extracted catalyst did not affect the reforming activity of this catalyst as will be described below. This finding suggests that the residual extractants can be removed from the catalyst at the high refox temperatures.

Example VII

The naphtha reforming activities of a redispersed Pt / Ir catalyst (containing 0.35% Fe) is compared in the example with a fresh Pt / Ir catalyst and one where the Fe is extracted. The results are shown in Table G. The reform reaction conditions are described in Example II.

TABLE G

Naphtha reforming activities of iron-extracted, alumina-supported __Pt / Ir catalysts_ (¾)

Catalyst Hours in ...,.

_ operation R0N (b) sca (A), fresh catalyst 72 103.1 176 10 143 102.2 143 215 101.5 124 335 101.8 134 455 101.6 126 (B), redispersed catalyst with 64 99 .9 91 ° '35% Pe 136 97.2 60 208 95.0 40 328 93.3 40 400 92.6 39 φ 20. (C), iron extracted from catalyst (B) 65 104.7 258 137 103/1 176 209 102.5 155 305 101.2 117 401 100.3 97 25 449 100.3 97 (a) see table F for full catalyst data (b) RON = research octane number (c) RCA = relative catalyst activity

After 400 hours of operation, the redispersed catalyst showed a reforming activity that was only about 30% that of a fresh catalyst. The reforming activity decline of the redispersed catalyst is an indication that iron contamination may be an important factor in the deactivation . After the 8303596 r? > * · 21 iron extraction is the reforming activity of. the extracted catalyst, after 400 hours of operation, about 80% of that of a fresh Pt / Ir catalyst. Thus, removal of iron from the redispersed sample provides a significantly more active catalyst.

Hst deleterious effect of 0.37% iron on the reforming activity of a Pt / Re catalyst is shown in Table H.

ïabsl h

Naphtha reforming activities of iron extracted, alumina supported

Pt / Re catalysts

Catalyst ^ Hours run RON ^ RCA ^ 10 (D), fresh catalyst 65 102.1 142 137 100.3 93 208 98.5 72 329 97.4 61 401 96.9 57 15 (E), iron doped catalyst. 63 99.2 80

Mt ° '37% FS 135 96.0 53 207 94.1 45 278 91.7 36 326 90.0 32 20 350 89.5 30 (F), catalyst (E) from extracted 64 102.5 155 heard iron 136 101f3 119 208 99.9 91 327 98.5 72 25 399 98.3 66 423 97.4 61 (a) see table F for full catalyst data (b) RON * research octane number (c) RCA «relative catalyst activity 30 De reforming activity of the iron-doped catalyst after 350 hours of operation is only 50% of that of a fresh catalyst.

The same iron-doped catalyst demonstrated a 50% lower hydrogen and carbon monoxide uptake than a fresh catalyst (see Table F).

8303596

+ VJ

22 f Thus, there is an excellent match between the catalytic ones

activity and the chemisorption measurements for iron-contaminated Pt / Re catalysts. The extraction of 80% iron from the iron-doped Pt / Re catalyst provides a catalyst that has hydrogen and carbon monoxide uptake and reforming activity close to that of a fresh catalyst. The correspondence between the chemical values and the catalytic reforming activities is an indication that iron poisons the bimetallic catalysts by interacting with the metal components. The fraction of the noble metals rendered inactive by iron 10 increases with increasing iron concentration.

This example demonstrates that inorganic acids are not equivalent to oxalic acid in terms of iron removal from catalysts. A catalyst containing 0/28% Pt, 0.27% Ir and 2.0% Fe on Al 2 O 3 -15 support was treated with hydrochloric acid and hydrochloric acid buffered with ammonium acetate, the results of which are shown in Table J.

TABLE J

Treatment ^ Recovered ^ Recovered catalyst Solution pH catalyst (wt% (wt% metals) ___________ _ _ Pt Ir Fe 20 50 ml 0.5 MHCl 1.0 83 0.25 0.26 1.6 50 ml 0, 25M 3.0 93 0.27 0.27 1.9 HCl / 0.25 ammonium acetate

(a) 3.0 g of catalyst (0.16 cm extrudates) was heated at 100 ° C

25 for 60 minutes.

(b) recovered catalysts were rinsed with five 100 ml parts of distilled water and dried at 120 ° C for 16 hours.

These results demonstrate that buffered HCl solutions give rise to large carrier losses without significant iron removal.

8303595

Claims (10)

A method of removing metal contaminants from a hydroconversion catalyst supporting at least one Group VIB, VIIB or VIII metal or containing an acid-resistant inorganic oxide, characterized in that the contaminated catalyst is contacted with a buffered oxalic acid solution. A method according to claim 1, characterized in that the oxalic acid solution is buffered by a buffering agent in an amount sufficient to maintain the pB in the range of 2 to 10. 3. Process according to claims 1 or 2, characterized in that the oxalic acid concentration is 0.0001 M to 5.0 M. 4. Method according to claims 1-3, characterized in that the temperature is about 0 ° C to 100 ° C. A method according to claim 1, characterized in that the carrier is acid-soluble. Process according to claim 1, characterized in that the catalyst contains at least one metal from the Co, Mo, Be or Ft supported on alumina, silica-alumina or zeolites. 7. Process according to claims 1-6, characterized in that the contaminated catalyst is contacted with a buffered oxalic acid for a sufficient period of time to extract the metal impurities without dissolving the inorganic oxide. 8. Process according to claims 1-7, characterized in that the oxalic acid solution is a mixture of oxalic acid in an aqueous / non-aqueous solvent system. 9. Method of removing metal contaminants from a catalytic cracking catalyst, which cracking catalyst contains at least one of alumina, silica, silica-alumina, zeolite or clays, according to the method in which the contaminated cracking catalyst is contacted with a buffered oxalic acid solution. 10. A method of removing metal impurities from a hydroconversion catalyst essentially as described herein with reference to the examples. 8303596