Patent Application
20190292680
This invention relates generally to coating bath solutions and more particularly to coating baths, e.g. electrolytes used in electrolytic deposition, and a method and system for removing contaminants from coating bath solutions.
Aluminum and aluminum alloys are used to create a wide variety of consumer goods and machine parts. Because aluminum and its alloys are very reactive and subject to corrosion it is desirable to apply coatings to the aluminum surfaces to provide corrosion, heat and abrasion resistance to the surfaces of these consumer goods and machine parts. The coatings may be formed from aqueous acidic solutions that dissolve some aluminum (and other metals) from substrate surfaces into the coating bath. In some processes the coating baths include materials that form water-soluble complexes with the aluminum, such that the aluminum does not form much sludge but remains in the bath. Particularly where coating baths contain aluminum, titanium and fluorine, the bath may build up aluminum-fluoride complexes to a level that interferes with coating. Coating baths that have this dissolved aluminum buildup include those used with or without the application of electromotive force.
Typically, these coatings may be applied by anodization, plasma electrolytic oxidation, micro arc oxidation, microplasma oxidation and similar processes. For example, one electrolytic coating process uses a bath containing dissolved components of complex fluorides and oxyfluorides of Ti, Zr, Hf, Sn, Ge, B, and mixtures thereof in the presence of phosphorous containing acids, phosphorous containing salts and mixtures thereof. Other dissolved components in the electrolytic coating bath can include: oxysalts of Zr, V, or Ti; salts of Nb, Mo, Mn and W; and alkali metal hydroxides or fluorides. This process is also often known as an electro ceramic coating process. In the present specification and claims the terms “anodization plasma process” and “electro ceramic coating process” refer to the same anodization process. In brief the process involves providing a cathode in contact with the anodizing solution, placing the article to be coated as the anode in the anodizing solution and passing a current through the anodizing solution at a voltage and for a time effective to form the coating on the surface. One can use direct current, pulsed direct current or alternating current with pulsed direct current or alternating current being preferred. Basic anodization plasma processes are disclosed in U.S. Pat. Nos. 7,569,132 and 8,663,807 which are whereby incorporated by reference. Most often the desired ceramic coating comprises titanium and/or zirconium oxide coatings. The anodization solutions have specific parameters that must be adhered to in terms of various contaminants and their concentrations need to be monitored.
When depositing a ceramic coating on aluminum parts using an anodization plasma process, aluminum (Al) gradually accumulates in the coating bath solution. In the present specification and claims the term aluminum parts refers to both pure aluminum parts and aluminum alloy parts unless noted otherwise, and likewise applies to parts coated with aluminum or its alloys. The aluminum that accumulates in the bath comes from the aluminum part as a result of etching that takes place during the anodization plasma process. Eventually the concentration of aluminum in the coating bath solution reaches a high enough concentration and causes the properties of the applied ceramic coating to deteriorate which in turn requires the bath solution to be disposed of. This leads to increased costs, waste and low efficiency of transfer of the desired ceramic onto the aluminum parts.
One problem discovered in attempting to remove dissolved aluminum-fluoride complexes from a coating bath containing dissolved titanium-fluoride complexes is that many separation methods do not distinguish between the two types of complexes. These separation methods remove both Al and Ti from the coating bath, which is an unacceptable result where Ti is a desired coating component. Prior attempts to selectively remove the aluminum by various single-step processes such as ion exchange, diffusion dialysis, crystallization or precipitation have all failed.
Thus, it would be desirable to create an economical, straight forward method and system for removing aluminum from electro ceramic coating bath solutions. Preferably the system and method could be implemented on existing coating systems with few changes to the process and most preferably it would function in a method that would not require shutting down of the electro ceramic coating line.
In general terms, this invention provides an economical system and method for removing aluminum from coating baths, e.g. an electro ceramic coating bath solution, and restoring the bath solution to a usable condition. This allows the bath life to be extended far beyond currently available life times. The process is highly effective and permits more efficient utilization of the coating composition. Although the process and apparatus are described with reference to removing aluminum from an electrolyte, those of skill in the art will recognize that the invention may be used for removing aluminum from other aqueous coating baths having similar chemistry.
An object of the invention is to provide a method for removing aluminum from coating bath, e.g. an anodization bath, solution comprising the steps of:
a) providing an aqueous acidic coating bath solution containing NH4+ and dissolved aluminum;
b) passing the coating bath solution through a strong acid cation exchange column in the Na+ form and exchanging Na+ for NH4+ in the coating bath solution thereby generating a first effluent and collecting the first effluent;
c) removing insoluble cryolite, Na3AlF6, solids from the first effluent of step b) thereby forming a supernatant effluent; and
d) passing the supernatant effluent from step c), after removal of the cryolite, through a strong acid cation exchange column in the NH4+ form and exchanging NH4+ for Na+ in the supernatant effluent thereby generating a second effluent and collecting the resulting second effluent.
The cryolite forms over a period of time and thus the effluent from step b) may be aged for a period time sufficient to allow for formation of the cryolite. Desirably, this period of time can be as short as 1 hour and can extend to 5 hours or more than 5 hours.
In some embodiments, the method further comprises a step prior to step a) of removing a selected amount of an aqueous acidic anodization solution containing NH4+ and dissolved aluminum from a bath or other source, providing the selected amount as the aqueous acidic solution of a) and after performing steps b)-d) described above, performing additional step e) returning the second effluent to the bath or other source and optionally f) replenishing the bath.
According to one aspect of the present invention (“Aspect 1”), a method is provided for removing aluminum from an anodization bath solution comprising the steps of:
Further aspects of the invention may be summarized as follows:
Aspect 2: The method of Aspect 1, wherein the anodization solution comprises water-soluble complex fluorides and/or oxyfluorides of elements selected from the group consisting of Ti, Zr, Hf, Sn, Al, Ge, B, and mixtures thereof; and wherein the anodization solution further comprises phosphorous containing acids, salts or mixtures thereof.
Aspect 3: The method of any one of Aspects 1 and 2, wherein the anodization solution in step a) contains more than 200 parts per million (ppm) of aluminum.
Aspect 4: The method of any of Aspects 1-3, wherein the cation exchange column in step b) comprises from 0.05 to 1.0 liters of cation exchange resin per gram of aluminum to be removed from the anodization solution from step a).
Aspect 5: The method of any of Aspects 1-4, wherein a rate of flow of the anodization solution through the cation exchange column in step b) is from 2 to 50 bed volumes per hour.
Aspect 6: The method of any of Aspects 1-5, wherein removal of cryolite in step c) comprises either filtering the cryolite from the effluent or separation from the effluent by a centrifugation process.
Aspect 7: The method of any of Aspects 1-6, wherein step c) further comprises aging the effluent from step b) in a tank for a period of time of at least 1 hour to allow for formation of the cryolite in the effluent prior to removing it in step c).
Aspect 8: The method of Aspects 7, wherein said period of time of aging the effluent from step b) comprises a sufficient amount of time to allow for formation of cryolite particles having a size of from 0.1 to 50 microns.
Aspect 9: The method of any of Aspects 1-8, wherein step d) comprises passing the effluent from step c) through the same cation exchange column as in step a), which was has been regenerated to be in the NH4+ form, to exchange NH4+ for Na+ in the effluent and collecting the resulting effluent.
Aspect 10: The method of Aspect 9, wherein the anodization solution in step a) is passed through the cation exchange column in a first direction and wherein the effluent from step d) is passed through the same cation exchange column in a counter current direction to step a).
Aspect 11: The method of any of Aspects 1-10, further comprising passing the effluent from step b) through a Na+ polishing ion exchange column prior to step c).
Aspect 12: The method of any of Aspects 1-11, further comprising passing the resulting effluent from step d) through a NH4+ polishing ion exchange column.
Aspect 13: The method of any of Aspects 1-12, wherein the cation exchange column in step a) is a different cation exchange column from the cation exchange column utilized in step d).
Aspect 14: The method of any of Aspects 1-13, further comprising the step of regenerating the cation exchange column from step a) to the Na+ form by passing a regenerating solution containing at least one regenerant salt selected from the group consisting of NaCl, Na2SO4, NaHSO4, Na3PO4, Na2HPO4, NaH2PO4, and mixtures thereof through the cation exchange column.
Aspect 15: The method of Aspect 14 wherein an equivalents excess of regenerant equivalents to cation resin equivalents is in a range of from 1-fold to 10-fold excess.
Aspect 16: The method of Aspect 14, wherein a flow rate of the regenerating solution through the cation exchange column is from 2 to 50 bed volumes per hour.
Aspect 17: The method of any of Aspects 1-16, further comprising the step of regenerating the cation exchange column from step d) to the NH4+ form by passing a regenerating solution containing at least one regenerant salt selected from the group consisting of (NH4)2HPO4, (NH4)H2PO4, (NH4)2SO4, (NH4)HSO4, (NH4)Cl, and mixtures thereof through the cation exchange column.
Aspect 18: The method of Aspect 17, wherein an equivalents excess of regenerant equivalents to cation resin equivalents is in the range of from 1-fold to 10-fold excess.
Aspect 19: The method of Aspect 17, wherein a flow rate of the regenerating solution through the cation exchange column is from 2 to 50 bed volumes per hour.
Aspect 20: The method of any of Aspects 1-19, wherein the resulting effluent after step d) has an aluminum content of from 200 to 3000 parts per million.
Aspect 21: The method of Aspect 14, further comprising a periodic step of regenerating the cation exchange column from step a) with HCl thereby removing precipitated cryolite from the cation exchange column followed by neutralizing with NaOH or NH4OH thereby regenerating to the Na+ form or NH4+ form.
These and other features and advantages of this invention will become more apparent to those skilled in the art from the detailed description of a preferred embodiment. The drawings that accompany the detailed description are described below.
Electro ceramic coating baths have stringent quality requirements that can only be met by maintaining the concentrations of various impurities below certain concentrations. One such impurity is aluminum which is etched from the aluminum parts and which accumulates at a rate of about 1.5-5.5 gm/m2 or more in the bath depending upon the processing parameters. At an aluminum concentration of about 1000-2500 ppm (1 to 2.5 gm/L) and higher in coating baths, e.g. electro ceramic coating baths, depending on the specific part or alloy, the coating quality deteriorates. At this point, the bath has historically been dumped and recharged with new solution resulting in increased raw material consumption and cost, downtime on high production lines, and waste disposal issues. With typical processes, even with optimized bath conditions, only about 30 wt. %, based on total weight of titanium from raw materials added to the bath, becomes part of the coating deposited on the parts. The process of the present invention selectively removes aluminum from the bath thereby indefinitely extending its life and preserving the expensive titanium in the bath for deposition on aluminum parts. By increasing bath life and keeping the raw material titanium in the bath, the inventive cleaning process can more than double the percentage of the titanium from raw materials that is deposited on the parts, meaning greater than about 60 wt. % of the titanium from the raw materials added to the bath will become part of the coating. Generally, about 40 wt. % of titanium added to the bath is lost with drag-out liquid that exits the bath with the coated part and sludge. As such, deposition of about 60 wt. % of the raw material titanium is use of nearly all available titanium for coating, considering drag out and sludge. The drag out can be reduced by various means that are not addressed herein.
A basic anodization process system, with which the method and system of the invention may be used, is depicted as a coating bath schematic in
The present process typically is a four or five step process for restoring an anodization bath solution having high concentrations of aluminum back to a working anodization bath solution with greatly reduced concentrations of aluminum, without removing significant amounts of titanium. As shown diagrammatically in
When using the inventive process, after IEX treatment, it is desirable to include in the process, a step to test the concentrations in the resulting effluent and/or the coating bath. Desirably one may test concentrations of the active components of the ceramic coating composition such as the complex fluorides and oxyfluorides of Ti, Zr, Hf, Sn, Ge and B and any other critical bath components, in particular those to be incorporated in the finalized ceramic coating on the aluminum part in order to achieve a return to good coating deposition.
The process can be run using a single strong acid cation exchange column 550 that starts out in the Na+ form and using a counter current flow through the column 550 (see for example
A schematic of a system for carrying out the inventive process utilizing two ion exchange columns 500 and 510 is shown in
As shown in
To reuse the resin in the IEX columns after a selected number of bed volumes or cleaning cycles, the IEX resin is regenerated. A Na+ salt is used to displace NH4+ from the resin used in the first IEX column 500 and a NH4+ salt is used to displace the Na+ from the resin used in the second IEX column 510. The exact choice of regenerating solution will depend on its cost, the potential for cross contamination of the bath with the counter ion, the ease of rinsing it from the resin bed, and the effectiveness of the actual cation exchange. In common IEX processes, the choices for the regenerants may be NaCl, to displace the NH4+ ions from the first column, and NH4Cl or (NH4)2SO4, to displace the Na+ ions from the second column. Applicants have found that the most effective NH4+ salts include the Cl−, hydrogen phosphates and SO4−2, with the Cl− rinsing most effectively from the column. However, both Cl− and SO4−2 are deleterious to some coating bath solutions, therefore more prudent regenerant choices may be those having counterions that are already part of the coating bath. For one embodiment of an electro ceramic coating process, regenerants may be Na3PO4, Na2HPO4, NaH2PO4, and mixtures thereof to displace the NH4+ from the first column and other regenerants may be (NH4)2HPO4, (NH4)H2PO4, and mixtures thereof to displace the Na+ from the second column, where phosphate is already part of the coating bath solution and so small amounts of carryover from the IEX column regeneration is unlikely to contaminate the bath solution.
A variation of this process is shown in
After precipitation of the cryolite, Na3AlF6, the contents of the aging/precipitation tank 600 are passed through a solids separation device 400, e.g. a filter and/or a centrifuge, subjected to filtering or centrifugation to remove the insoluble cryolite, forming a supernatant effluent. This completes the first half of the cleaning process. The second half of the cleaning process is then begun by taking the filtrate or supernatant from the centrifugation, i.e. supernatant effluent, and passing it, preferably downwardly, through the primary IEX column 550 thereby exchanging NH4+ in the resin (derived from the coating bath in the first pass upward flow) for Na+, thereby producing a second effluent. The second effluent may then be passed through a smaller NH4+ polish column 530 and finally the second effluent is returned to the coating bath tank 200 or sent to another reservoir, such as surge tank 450 shown in
One advantage of this embodiment shown in
In carrying out the present invention, the aluminum concentration in the coating bath solution is preferably maintained at a low, non-zero amount; as the bath is used aluminum builds up from amounts of as low as 1 ppm and will continue to increase to amounts of 4000 ppm and higher due to dissolution of Al from the substrate during coating deposition. Eventually, the aluminum in the coating bath begins to affect coating quality. Through use of the instant cleaning process, Al concentration may be maintained in a range of at least about, 100, 120, 140, 150, 160, 170, 180, or 190 ppm to no more than about 4000, 3500, 3000, 2500, 2000, 1500, 1000, 500, 250 or 200 ppm. Desirably the aluminum concentration in the working coating bath is maintained at about 200 to about 3000 ppm, more preferably from about 500 to about 1500 ppm.
The size of the IEX resin bed depends on the expected amount of aluminum to be removed from the liquid feed from the coating bath. The IEX resin bed size may be adjusted to a range of about 0.05 to 1.0 liters of resin per gram of aluminum to be removed, more preferably from 0.15 to 0.35 liters of resin per gram of aluminum to be removed. Desirably, the IEX resin bed size can be at least about 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, 0.10, 0.11, 0.12, 0.13, 0.14, or 0.15 liters of resin per gram of Al to be removed and at least for economy and for a smaller apparatus size, not more than about 1, 0.75, 0.50, 0.40, 0.35, 0.30, 0.25, 0.20 liters of resin per gram of Al to be removed.
The flow rate of the feed into the IEX columns may range from 1 to 50 bed volumes per hour and more preferably from 5 to 25 bed volumes per hour. Desirably, the feed flow rate is at least about 0.5, 1, 2, 3, 4, 5, 6, 7 or 8 bed volumes/hour and not more than about 50, 45, 40, 35, 30, 25, 20, 15 or 10 bed volumes/hour.
Particle size to be removed by the filtration or centrifugation may range from 0.1 to 50 microns, more preferably from 0.5 to 20 microns.
As described above, the present process utilizes as the primary IEX column 550, a resin that is a strong acid cation exchange resin that starts in the Na+ form; alternatively, the process may utilize a first column of resin that starts in the H+ form to remove NH4+ and a second column of resin in the Na+ form to remove H+. Suitable examples of strong acid cation exchange resins in H+ or Na+ form include Amberjet 1200H, Dowex G26H, Dowex Marathon 650C, Dowex Marthon C, commercially available from The Dow Chemical Co.; Lewatit MonoPlus S 108H, 200 KR and 215 KR, commercially available from Lanxess Aktiengesellschaft; Diaion™ PK208, SK102, SK104, SK1B, SK110, SK116, UBK08, UBK10, UBK12, UBK16, commercially available from Mitsubishi Chemical Corporation; and Purolite C100X16MBH, C160H, C100, C100E, C100X10, C120E, Puropack PPC100H and Purofine PFC100, commercially available from Lenntech BV. Porous or gel matrix resins with a matrix that is styrene-divinylbenzene (DVB) base with sulfonic acid or similar functional groups may be used. Strong acid cation exchange resins in the Na+ form are preferred.
The flow rate of the regenerant through the IEX columns may range from 1 to 10 bed volumes per hour, more preferably from 2 to 5 bed volumes per hour. The number of equivalents of cations in the regenerant solution per equivalent in the resin bed may range from 1 to 10, more preferably from 1.5 to 3.0. Desirably, the number of equivalents of cations in the regenerant solution per equivalent in the resin bed may range from at least about 0.5, 0.75, 1.0, 1.25, 1.5, 1.75, 2.0, 2.25, 2.5 or 2.75 equivalents of regenerant cations per equivalent of resin bed cations and at least for economy not more than about 15, 13, 11, 10, 9, 8, 7, 6, 5, 4 or 3 equivalents of regenerant cations per equivalent of resin bed cations.
Description is made herein with reference to anodization bath solutions but will be understood by those skilled in the art to apply to other coating baths having comparable chemistry or components and/or dissolved Al to be removed from the coating bath.
The cryolite precipitation process was not the first process evaluated to remove Al from anodizing coating baths containing Al, Ti and other various other components. Earlier attempts to separate the Al are briefly described below:
Cation exchange with various types of resins (i.e. strong acid cation (SAC), weak acid cation, iminodiacetate, aminophosphonic acid, etc.) were tested, but none exhibited any capacity for Al since the Al bond energy in the fluoroaluminate anion (i.e. AlF6−3) that exists in solution is far greater than that of any Al species that might bind to the functional group on the IEX resin. Experiments often give the impression that some Al is removed but careful mass balances showed the drops in [Al] in these experiments were due to the water of dilution from the resin beads when relatively large resin-to-solution ratios were tested in batch experiments.
Anion exchange with various types of resins (i.e. strong base anion, moderate base anion, weak base anion, etc.) was tested. However, all these resins had a greater affinity for the fluorotitanate anion (i.e. TiF6−2) in solution than the fluoroaluminate anion. As such, all anion exchange processes removed more fluorotitanate than the fluoroaluminate making direct purging of the bath to drain more efficient. Also, both the fluoroaluminate and fluorotitanate anions precipitate at pH's>˜4. As such, regenerating an anion resin proved difficult to do without precipitation occurring inside the resin beads, which is undesirable.
Low temperature precipitation was attempted by cooling the used coating bath to −5° C. and crystals that formed were slightly enhanced in the ammonium fluoroaluminate compared to the ammonium fluorotitanate but not enough to justify further work.
Salts were precipitated from the bath by tiny, e.g. dropwise, incremental additions of 1% NH3. The least soluble species in this experiment was found to be Ti3(PO4)4 which would precipitate first, which data is consistent with the EDAX of precipitate in the working coating baths. Precipitation of Ti from the bath is undesirable, since this is a primary element in the anodic coating bath tested.
Seventeen resins of 9 different types/functionalities were tested for viability in removal of Al without accompanying loss of Ti. In this experiment, 5 mls of anion resins or 10 mls of cation resins were mixed with 120 mls of a used anodic coating bath containing ˜470 ppm Al and ˜4980 ppm Ti. After shaking the samples and letting them sit overnight, decanted the liquid and submitted it for Inductively Coupled Plasma (ICP) analysis. Results are shown below in Table A.
The above Table A shows the % of components which were removed and the relative selectivity of each resin for Al/PO4 and Al/Ti. The Al/Ti selectivity indicates only cation resins had selectivities>1. A selectivity of 1 represents what can be achieved by simply bleeding the bath (i.e. equal % drops for both Al and Ti). Only 3 of 17 resins had selectivity>1. The resins having the greatest selectivity for Al were then investigated in column experiments.
Tested the five resins listed in Table B 1, below, in 10 ml columns. All resins were converted to the H-form prior to being used.
Pumped used anodizing bath (470 ppm Al and 4980 ppm Ti) as feed through the columns to saturate the resin. Regenerated columns with ˜5 BV's of HCl and analyzed regenerant effluents by ICP, results are shown in Table B2, below.
Results in Table B2 showed very little Al and only slightly higher amounts of Ti in the regenerants. Also noted that some tramp metals (e.g. Ni, Zn, etc.) appear in higher quantities in the regenerants than the used anodizing bath, especially those from the iminodiacetate (IDA) columns.
Tested the four resins listed in Table C, below, in 10 ml columns. All resins were in the Na-form prior to being used.
Pumped through ˜70-80 BV's of used anodizing bath as feed (470 ppm Al and 4980 ppm Ti), collected seven cuts of effluent from each column which were then analyzed for Al (by UV-absorbance). Regenerated columns with 10% HCl and analyzed regenerant effluents by ICP and NH3 electrode. Results are shown below in Table C2.
Found insignificant amounts of Al or Ti in regenerant cuts but did find NH3 in both SAC regenerants—enough to saturate the beads. The data above shows that the SAC resins in the Na-form pick up NH4+ very effectively—even at pH ˜2.5.
Tested five resins listed below in Table D1, as H-form in 59-73 ml BV columns, carefully quantifying water of dilution in the system.
Analyzed multiple effluent cuts for Al by UV-absorbance method. The water of dilution was mathematically factored out of the Al concentrations, and for the H-form IEX resins, Al concentration results show no Al was adsorbed on H-form IEX resin beads.
In a first example showing use of the present invention, a tote of used electrolyte coating bath solution was used as the feed for the bath cleaning system. The used coating bath solution contained 1560 ppm of aluminum and 5730 ppm titanium in solution, both being complexed with fluoride. The suspended solids in this feed were rich in Ti, P, and O as determined by Energy Dispersive X-Ray Spectroscopy (EDAX) and these suspended solids were removed by filtration through a 1-micron filter followed by centrifugation such that the feed to the actual experiment was about 1 nephelometric turbidity units (NTU) compared to 400 NTU for the starting material from the tote. This was done so that there would be no ambiguity whether solids present at the end of the experiment came from the feed or were produced during the experiment.
One hundred milliliter (ml) aliquots of this feed were mixed in four plastic bottles with 5, 10, 15, and 20 ml, respectively, of strong acid cation exchange resin in the Na+ form, commercially available from Mitsubishi Chemical Corporation. Almost immediately, a white precipitate formed in each bottle. The bottles were intermittently shaken during the day, but no mixers were used. After sitting overnight, the solids, both dark resin beads and a white powder, were removed from the samples by centrifugation. The resin beads were separated from the white powder by filtration through a 100-micron polypropylene monofilament screen. Then the recovered liquid was analysed as described below.
The initial feed and the recovered fluids from the four IEX treated products were analysed by a battery of tests: pH; aluminum content by a UV-absorbance method as described in Journal (American Water Works Association), Vol. 82, No. 5, pp 71-78 (May 1990); coating bath contents by UV-absorbance method; free fluoride concentrations by a fluoride selective electrode; detection of undissociated HF by a Lineguard 101 meter; Inductively Coupled Plasma (ICP) testing for detecting Al, Ti, Si, Na, K, and PO4; total fluoride analysis by a Huckabay distillation; and measurement of NH3 by an electrode to determine NH4+ concentrations. The measurement of NH3 is taken as the concentration of NH4+ in the sample. Below in Table 1, the results for the ICP analysis and total fluoride are reported. The concentration results presented in Table 1 are corrected for the amount of dilution water added with the resin additionally the % recovery from the feed concentration is presented for each component.
The results show the benefits of the current process. The amount of aluminum left in the recovered liquid was dramatically reduced showing that the resin was able to remove aluminum from the bath solution. Almost 80% of the aluminum was removed by the 20 ml resin sample. The loss of aluminum in the various samples closely corresponds with the loss of NH4+ in the same samples due to the ion exchange reaction. In addition, one sees a dramatic rise in the Na+ concentrations as expected since it is being displaced from the resin by the NH4+. Importantly, the ion exchange resin did not remove most of the other main components of Ti, Si, K, PO4, and total fluoride. These were all retained at a concentration of about 70% or greater as compared to the original feed. The amount of Ti was retained at better than 85% which is significant for repeated uses of the cleaned bath solution. Losses of Ti and PO4 are small and appear to be due to interstitial liquid not recovered from centrate when the cryolite was removed.
In Table 2 below are present the values for total milligrams (mg) in the solutions, millimoles in solutions (mM), change (Δ) of moles in the solutions after passing through the resins, and the mole ratio of ΔF/ΔAl in the solutions. A mass balance analysis of the results from ICP and total fluoride shows that the molar ratio change of F to Al is consistent with precipitation of Na3AlF6 since the molar ratio of Δ F to Δ Al is approximately 6 in all the samples. In addition, EDAX analysis of the solid precipitates, which were largely cubic crystals of 4 to 6 microns, showed they were rich in Na, Al, F. The precipitates also contained a little Si. A scanning electron micrograph picture of the precipitate from the 10 ml of resin sample from this example is shown in
Example 2 used the same coating bath feed and preparation as described in Example 1. In this experiment, two small IEX columns made from clear Schedule 40, ½″ PVC pipe, inner diameter of 0.608 inches by 18 inches tall were filled with 66 mls of a strong acid cation resin in the Na+ form, commercially available from Mitsubishi Chemical Corporation. This produced a resin bed length of approximately 14 inches. The resin beds were rinsed with reverse osmosis (RO) water until their effluent conductivities were <20 micro Siemens/cm. The water level in each column was then drained down to approximately ¾″ above the resin beads before starting the feed so as to minimize the amount of dilution in the first effluent samples. The coating bath feed was pumped in a downward flow through each column at approximately 7.7 mls/min, which was equal to 7 bed volumes (BV)/hour. A total of 12 samples of each column effluent were collected at 7-minute intervals. After feeding, the columns were rinsed with RO water at the same flow rate for 15 minutes.
Precipitates formed in most of the effluent samples, almost immediately for the early samples and overnight for the later samples. This indicates that the precipitation process will require some time to occur. The solids in the first sample were fluffy while the solids in later samples were more “chippy”, e.g. larger firmer particles. Prior to analyses listed below, the solids in all of the samples were removed by centrifugation. A total of 28 samples, feed+13 effluent samples, from each column were analysed by the same tests as in Example 1.
The volumes of feed and resin in Example 2 were chosen such that the resin bed should have been exhausted, all of the Na+ in the resin beads replaced by NH4+, at about 10 BV's of feed assuming 100% efficiency of exchange. The same data set generated a mass balance analysis which indicated that the net process removed about 5.7 grams of Al per liter of resin.
The graph in
After rinsing, one column was regenerated with 5 BV's of 10% HCl followed by neutralization with NaOH and the other column was regenerated with 5 BV's of 15% NaCl to see if one mode of regeneration is more efficient than the other. The flow to each column was approximately 2.8 BV/hour. The sample bottles were weighed before and after collection so an accurate mass balance could be made. The analytical results from the regenerant samples showed that the NaCl regenerant performed as calculated, providing satisfactory results close to theoretical and good rinseability from the IEX column. The HCl regenerant analysis showed total cations removed were higher than theoretical. The data suggests that some cryolite was precipitated inside the IEX column, which was then dissolved by the HCl regenerant.
The main purpose of this example is to complete all steps of the bath cleaning process and then use the cleaned bath to coat test coupons to confirm usefulness of the process. To do this, a larger IEX column was needed so the height of the IEX column was increased to 67 inches. It was filled with 250 mls of a strong acid cation resin in Na+ form. This produced an IEX resin depth of about 53.5 inches. The experimental protocol was similar to that followed in Example 2. Approximately 15 BV's of used coating bath feed were pumped in a downward flow through the column at a rate of 47 mls/min, approximately 11.3 BV/hour. Thirteen samples of effluent were collected at six-minute intervals.
The column was then rinsed with RO water for 15 minutes and then regenerated with 6 BV's of 10% NH4+ Cl pumped at 3.3 BV/hour. This amount of regenerant was in excess. The effluent samples were analyzed by UV-absorbance method.
The graph in
Since each effluent sample was about 280 mls, the cryolite was separated by filtration on #410 filter paper, which has a nominal rating of 1 micron. The filter cakes were not washed but were allowed to air dry for several days. For effluent samples #1 through #8, the cumulative weight of the Al removed was calculated to be 1.33 gm while the measured weights of the air dried filter cakes summed to 11.54 gm. This gives a weight ratio for the cakes to removed Al=11.54/1.33=8.66, which is quite close to the molecular weight ratio of cryolite to Al, 209.94/26.98=7.78, giving more evidence that the precipitate is cryolite, or contains mostly cryolite, with a small amount of occluded mother liquor.
Three samples of liquids from the IEX cleaning process were tested for coating quality: A) a portion of the original feed of used coating bath filtered to remove solids; B) a coating bath after the first IEX to Na+ form generated a first effluent and filtration of the first effluent to remove cryolite to make the supernatant effluent which was used as the coating bath; and C) a coating bath after the second IEX changing from Na+ form back to NH4+ form plus minor replenishment to correct for water of dilution from the IEX treatment. The three sample coating baths were used in an anodization process to coat small coupons of aluminum alloys 413 and 6061 using identical electro ceramic coating parameters as known in the art.
Both first and second IEX treatments were performed using identical flows, 47 mls/min=about 11.3 BV/hour. A total of 6 effluent samples were collected at six-minute intervals.
With this slightly replenished bath, small coupons of aluminum alloys 413 and 6061 were coated using standard electro ceramic coating parameters. The coating thicknesses was determined by a Portaspec wavelength dispersion X-ray fluorescence spectrophotometer and the results are summarized in Table 3 below. Performance is summarized with indicators as follows: “X” indicates unacceptable coating weight; “O” indicates low coating weight; and “+” indicates excellent coating weight.
The data clearly show that the bath in the Na+ form gives an unacceptable coating which is consistent with expectations. The cleaned bath was able to produce very good coatings having very good coating thicknesses. These results demonstrate the usefulness of the present process for cleaning and restoring coating bath solutions.
The foregoing invention has been described in accordance with the relevant legal standards, thus the description is exemplary rather than limiting in nature. Variations and modifications to the disclosed embodiment may become apparent to those skilled in the art and do come within the scope of the invention. Accordingly, the scope of legal protection afforded this invention can only be determined by studying the following claims.
| Number | Date | Country | |
|---|---|---|---|
| 62440024 | Dec 2016 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/US2017/068658 | Dec 2017 | US |
| Child | 16438988 | US |
