Title | Alkaline Hydrolysis Resistant and Durable Refractory Linings for Fired Process Heaters |
Creator | Moody, K.J. |
Date | 2017-12-12 |
Description | Paper from the AFRC 2017 conference titled Alkaline Hydrolysis Resistant and Durable Refractory Linings for Fired Process Heaters |
Abstract | Since the late 1960's there have been problems associated with rapid lining deterioration of insulating castables used in Fired Process Heaters. New ceramic compositions, using patented hydrophobic insulating aggregates, eliminate alkaline hydrolysis reactions in the refractory linings of Fired Process Heaters. This hydrophobic technology also enables these new compositions to be fired in without holds and without dryout and at rates of 200oF to 400oF in new installations and in turnarounds. ; Fired Process Heater modules often wait in storage from 12 to 18 months before erection and completion of an operable refinery - an invitation to Alkaline Hydrolysis deterioration of the insulating castable lining. Expensive dryout procedures have been the common method to partially mitigate these deteriorations. To this end, API 560 Section 11.5.1 prescribes and requires drying out the lining within 45 days of installation at temperatures of 500oF (260oC). Dryout procedures only delay the Alkaline Hydrolysis Reaction, they do not eliminate the potential reactions nestled in the matrix of the insulating castables. ; However, experience proves that too often the risk and costs of alkaline hydrolysis and related downtime is significant to Owners even with dryout. The solution is to "prevent the disease" not to "cure it". Serious gaps exist in the support provided by supplying vendors and manufacturers to avoid these problems. Until now, Owners are left without a dependable alternative or a reasonable level of accountability and have typically paid for the consequences of the problems from Alkaline Hydrolysis of shutdown and delay.; These new technologies, using patented hydrophobic insulating aggregates and high purity raw materials, are engineered to prevent the Alkaline Hydrolysis Reaction and eliminate deterioration of the lining compositions. This paper describes the new ceramic compositions using hydrophobic aggregates, the chemistry of the reactions, and rapid test methods that demonstrate how these new technologies do "prevent the disease" of alkaline hydrolysis and enable fast firing of the installed linings. |
Type | Event |
Format | application/pdf |
Rights | No copyright issues exist |
OCR Text | Show Alkaline Hydrolysis Resistant and Durable Refractory Linings For Fired Process Heaters Kenneth J. Moody, BNZ Materials, Inc Executive Summary Since the large scale conversion from insulating firebrick to castables in the late 1960's there have been problems associated with rapid lining deterioration of insulating castables used in Fired Process Heaters. The rapid deterioration of the insulating castable lining is due to uncontrolled alkali reactions with the calcium aluminate binder. Various methods have been tried over the years to mitigate this rapid deterioration of the insulating castable lining. One of those methods is to dryout the lining to 500oF (260oC) within 45 days of installation as prescribed by API 560 Section 11.5.1. The dryout delays the reaction only if the Fired Process Modules can be kept 100% dry while waiting for erection and start-up of the refinery. This length of time can be from 12 to 18 months. This long term storage of the Modules in the presence of any moisture provides time for the Alkaline Hydrolysis reaction to take place and damage the insulating castable lining. Project experience has proven that the risk of Alkaline Hydrolysis is significant to the Owner with or without dryout. Historically there has been a serious gap in the support provided by the supplying vendors and manufacturers to avoid these problems. For years owners have been frustrated at the lack of dependable costeffective alternatives, or a reasonable level of accountability from refractory producers and contractors from a project execution basis. The Owner has typically paid for the consequences of the problems. Repair of damages can be extremely costly, causing one refiner to pay $4 MM to repair damage caused by Alkaline Hydrolysis. Significant innovation in the engineering of insulating castables used in Fired Process Heaters has led to compositions using patented hydrophobic insulating aggregates and high purity raw materials that control the Alkaline Hydrolysis Reaction and eliminate deterioration of the lining compositions. These proprietary compositions, using patented hydrophobic aggregates, create a compositional framework where optimum casting and gunning properties can be designed. These 1|P a ge compositions reduce the water demand for installation by 50%, maintain the density and strengths required by the application, significantly lower firing shrinkage, and enable the composition to be dried out at rapid rates without weakening the castable lining. These new technologies provide the Fired Process Heater engineers and end users with solutions that are durable and cost effective. Background There are different designs of heaters built to meet the process requirements, but all of these fired heaters use an insulating refractory to effectively capture and transfer the heat generated into the oil feed. Of the various types of insulating refractories used in fired heaters, this paper will concern only the cement bonded insulating castables that are installed by casting or gunning. For the past 60 years there have been problems associated with rapid lining deterioration of insulating castables used in Fired Process Heaters. Most if not all of these lining failures actually happen before the Fired Heater has been installed in the refinery/petrochem process. History has shown that lightweight castables with a fired density below 75 lbs/ft3 (1201 kg/m3) have had this method of failure when a short time has passed after installation and before the heater has been put into service. If the fired heaters were placed into process within 2 to 4 weeks after castable installation and were fired into operation temperatures, the lining deterioration is minimal. However, the main Hydrocarbon processing companies have determined that there are logistic/costing benefits to have many of these larger refinery and petrochemical processes broken into modules that are fabricated with refractory installed in areas far from the refinery/plant. After these modules are fabricated and the refractory installed they are then shipped, usually in sealed plastic to the final refinery destination to be assembled. Logistics and changing refinery scheduling mean that it often takes several months and sometimes even more than a year for these modules to arrive at the refinery location and get commissioned. The delays from installation to start-up can go from days and weeks to months and years. This created a new set of problems for the end user, the installer and refractory manufacturer. These longer delays, as will be presented, often results 2|P a ge in an almost perfect environment where the natural effects and reactions of the calcium aluminate hydrate along with the normally available alkalis, moisture, CO2, and temperature (before, during and after installation), created a reaction that would destroy the calcium aluminate bond and at the same time would be self-generating. These chemical reactions are called Carbonation and Alkaline Hydrolysis where the alkalis and CO2 transported by moisture react with the calcium aluminate hydrated binder. This reaction destroys the bond which holds the castable together. The reaction typically starts at the surface but in some cases within the castable. Conventional Castable Design The refractory linings for fired heaters can use insulating castables and ceramic fibers depending on the fired heater design and application requirements. Listed below are the typical minimum selection criteria for fired heater castables with the key properties of density, compressive strengths, permanent linear change and use temperature. Thermal Conductivity, typically controlled by the installed density, is in the range of 1.8 BTU·in/hr·ft2·oF (0.26 W/mK) to 2.01 BTU·in/hr·ft2·oF (0.29 W/mK) at 800oF (427oC). Table 1 Minimum Selection Criteria for Fired Heater Castables Engineering Specifications Dry Installed Density, lbs/ft3 (kg/m3) Exxon Mobil/BP ≤65 (≤1040) Minimum Compressive Strength, psi, (MPa) 500 (3.4) Permanent Linear Change %, Maximum Use Temperature Limit, oF (oC) -0.5/-0.4 ≥2300 (≥1260) These lower density and Thermal Conductivity value specifications drive the refractory manufacturers to develop castables with lower densities. In order to achieve these densities other super lightweight aggregates are used by some producers in addition to the standard expanded clay/crushed insulating firebrick aggregates. These aggregates, like perlite and vermiculite, bring chemistries which include soluble alkalis like sodium, potassium, and magnesium. These alkalis by themselves are not generally harmful, but in combination with long term storage 3|P a ge with calcium aluminate hydrates, they can exacerbate the Alkaline Hydrolysis Carbonation reaction. Listed below in Table 2 are the various lightweight aggregates currently being used to manufacture insulating castables for fired heaters. Table 2 Various Insulating Aggregates Used in Castables for Fired Heaters Insulating Aggregates Chemistry, Wt.% Al2O3 SiO2 Fe2O3 TiO2 CaO/MgO K2O/Na2O Bulk Density Loose Fill, lbs/ft3 ( kg/m3 ) Service Temperature, o F o C Perlite Vermiculite Expanded Clays Insulating Firebrick (Anorthite Mineralogy) Insulating Expanded Firebrick Shale (Mullite (Haydite) Mineralogy) 19.5 70.0 0.8 0.1 0.3 8.2 11-16 38-46 5-13 1-3 1-3 3-7 27.1 64.3 2.1 2.0 0.8 3.3 38 45 0.3 1.6 15 0.5 33.4 62.9 1.1 1.2 0.6 0.6 24 63 5.5 1.5 4.0 2.0 6-11 ( ) 4-8 ( ) 28-35 ( ) 28-35 ( ) 26-35 ( ) 50-65 ( ) 1500 (871) 1500 (982) 2700 (1482) 2400 (1315) 2500 (1371) 2190 (1199) The combination of perlite with any of these other lightweight aggregates do produce castable with densities in the proper range for fired heater applications. Different levels of perlite would be required with the different denser aggregates to meet the installed density and would inherently bring in to the composition significant quantities of soluble sodium and potassium. The soluble alkalis that these aggregate combinations bring to the composition promote the alkali hydrolysis/carbonation reaction. The resulting effects of this higher alkali content increases the fired strengths by forming low melting glasses during the initial firing and when cooled down to room temperature makes a strong glassy structure. The cold crushing strength test for monolithics is done at room temperature and not at operational temperature of 1350oF to 1850oF. Due to the glassy nature of these 4|P a ge compositions, the cold strengths tested at room temperature are not seen at the operational temperatures. However, the resistance to hot load at temperature has been reduced as compared to low alkali compositions. In addition the high levels of soluble sodium creates a supply chain of key reactants that promotes the Alkaline Hydrolysis/Carbonation reaction. Chemical Reactions and Testing During a six year investigation on these reactions, there have been several factors identified that individually and in combination contribute to the development of Alkaline Hydrolysis. In each case, the raw materials promoting this reaction are experimentally removed and evaluated to best understand their role. In Appendix 1 there are several basic chemical reactions. First is the Calcium Aluminate Cement Hydration reaction, second is the Carbonation reaction, and third the Hydrolysis reaction. In combination these three reactions create the foundation for the destruction of the calcium aluminate binder and weakening the castable so that it is no longer meeting the specifications for the fired heater insulation. The high porosity of these low density castables result in high water levels in the castables and promotes the reactions since this excess water allows the percolation of the carbonate ion (CO32-) and movement of the sodium ions toward the Calcium Aluminate Hydrate phases. In addition, humidity and mechanical moisture are key to the promotion of the reactions providing transport for the reactants and reaction products. Hydration In hydration water molecules enter the mineral lattice forming a hydrate form of the original raw material. A good example is the reversible change between anhydrite and gypsum. This is where the Calcium Sulfate Anhydrite reacts with water and forms Gypsum, the hydrated form of Calcium Sulfate. The calcium aluminate cement reacts in a similar way and forms several kinds of hydrated forms based on the ambient temperature. Just as with Gypsum, if heated, it will transition back to the anhydrite form, so it is with these calcium aluminate 5|P a ge hydrates. If they are partially dried out they will reverse back to the hydrate form when moisture is available. When the castable is partially dried out, some of the water molecules are removed by dehydroxlyzation. Once some of these water molecules have been removed, there remains unsatisfied charges in the cement hydrate phases. These phases become hygroscopic to satisfy these charges, and act much like an activated adsorbent attracting moisture that is in the air as humidity or by infiltration by rain. If the castable lining is partially dried out, it MUST be kept dry, or Alkaline Hydrolysis/Carbonation will start up again. This is shown in Appendix 1 Section A. Carbonation Carbonation is a complex reaction which causes Calcium Aluminate Hydrates to be dissolved through the reaction as shown in Appendix 1, Section B. For the Carbonation reaction to start, there must be soluble sodium ions present in the moisture in the castable. The sodium ions come from the aggregates like perlite, and from clays and other typical ingredients in the insulating castable. Secondly, the dissolving of CO2 into the water creates CO32- and continues the reaction by increasing the solubility of sodium into the water and thereby increasing the rate of the Carbonation and Alkaline Hydrolysis reaction. In some instances fired heater castables surfaces have been coated to reduce the amount of CO2 for the reaction. This only postpones the inevitable carbonation reaction because it does not stop the primary driver of sodium ions in the moisture. These two reactions, providing Sodium ions and Carbonate ions, create a reaction with the Calcium Aluminate Hydrate which, over time, dissolves the hydrate in the base stable forms of Calcium Carbonate and Alumina gel. Hydrolysis The most significant weathering reaction is Hydrolysis, in which minerals break down by reacting with water and carbonate ions, CO32-. Basic geology provides some interesting insight into these reactions. Carbonate ions and water produce clay minerals from the Hydrolysis of feldspars, like the Orthoclase mineral family 6|P a ge and the Plagioclase mineral family. Hydrolysis reactions with these types of minerals can be arranged in order of their resistance as follows from the less stable to the more stable. Mullite and silica minerals are not affected by Hydrolysis reactions. However, there are some clays which contain the mineral albite which easily reacts to the Hydrolysis reactions and tends to promote these reactions as seed pockets in the castable formulation. It has been found that Anorthite has a greater potential to react to Hydrolysis than albite. Shown below is a chart demonstrating the increasing stability of various minerals to the Hydrolysis reaction. In Appendix 1 Section C are the chemical reactions illustrating how Albite and Anorthite, two primary minerals found in some fired heater compositions, are dissolved by this reaction into kaolinite and quartz. Table 3 The Increasing Stability of Minerals under Hydrolysis Reactions Minerals Increasing Stability Olivine Anorthite Pyroxene Hornblende Albite Biotite Orthoclase Muscovite Quartz Mullite Kaolinite In addition to these hidden discoveries of minerals that are dissolved with water and carbonate ions, there has been discoveries regarding how the purity of the calcium aluminate cements can also have a significant effect on the rate of reaction with minimal quantities of sodium ions. The high purity cements have the following hydrate reactions after mixing with water. These cements contain silica, iron and other impurities in the range of less than 0.4 weight percent of the cement. 7|P a ge Table 4 Calcium Aluminate Hydrates Formed from High Purity Cements Temperature of Curing Hydration Products o <50 F CAH10 + AHx (gel) 50oF to 80oF CAH10 + AHx (gel) C2AH8 + AH3 (gel) >80oF C3AH6 + AH3 (crystal) C=CaO; A=Al2O3; H=H2O These phases can react with soluble sodium and carbonate ions, but when the soluble sodium is strictly controlled the reaction can be eliminated. During the first year of testing the new patented hydrophobic coating, much of the work used intermediate and low purity calcium aluminate cements. These low purity and intermediate purity cements have many other different phases formed and will partially hydrate. Listed below are the typical chemical analysis and phases seen in all three qualities of cements. Table 5 Calcium Aluminate Phases Present in Low and Intermediate Purity Cements Cement Type Chemistry Al2O3 Fe2O3 CaO SiO2 Phases Present in As Received Cement Fast Hydration Low Purity Intermediate Purity High Purity 39 - 50 % 7 - 16 % 35 - 42 % 4.5 - 9 % 55 - 66 % 1-3% 26 - 36 % 3.5 - 6.0 % 70 - 90 % 0.0 - 0.4 % 9 - 28 % 0.0 - 0.3 % CA CA C 4A3Ṥ C12A7 C12A7 C C Slow Hydration CA2 CA2 C 2S C 2S C4AF C4AF C2AS C2AS Non-Hydrating CT CT A A C=CaO; A=Al2O3; S=SiO2; Ṥ=SO3, F=Fe2O3; T=TiO2 CA C12A7 CA2 A 8|P a ge It was seen from accelerated AH/Carbonation tests with soluble sodium free compositions using low purity and intermediate purity cements that there were clear indications of carbonation and hydrolysis reactions that destroyed the cement phases. The phases associated with the low and intermediate purity calcium aluminate cements, as shown above, demonstrated reproducible Hydrolysis reactions where little or no soluble sodium was present. These reactions tended to show expansion in the middle of the samples along with surface effects. The same formula substituting high purity cements for the low and intermediate cements eliminated the reaction. The pictures below show the changes from the starting of the test until the castable had significant reaction on the surface and in the internal matrix. Formula 23 C was the favored formula after all of the key physical property testing, but failed in the rapid AH/Carbonation test. Picture 1 Comparison of Formula 23 C Containing Intermediate Purity Cement with No Soluble Sodium Formula 23 C One Week after CO2 Bath Formula 23 C after 7 Weeks in AH Test Test Development for AH/Carbonation and Hydrolysis Detection To determine which raw materials or combination of raw materials are susceptible to these reactions, an accelerated test was developed. The primary factors of driving the reactions are; soluble sodium, high levels of carbonate ions (CO32-), high levels of moisture in and around the samples, temperatures from 50oF to 120oF (10 to 49 oC), and high volumes of water moving through the castable. The testing procedure was to cast or hand ram the samples into 4.5" cubes, cure for 24 hours, then place in a water bath that had CO2 bubbling through the water for 7 days. These cubes were fully submerged in the CO2 bath. After the week immersed in 9|P a ge water, the samples were placed in aluminum pans half filled with water so that they were half in the water and half out of the water to allow high volumes of water to move through the castable matrix. The pans were partly covered allowing evaporation which pulled the water in the pan through the samples. This procedure was designed to pull any and all soluble alkalis through the samples and promote rapid Carbonation or Hydrolysis. The evaporation would percolate the water from the pan through the castable into the air creating a strong driving force to extract any sodium ions that might over time be leached out of the mineralogy of the composition. The need for this accelerated test is to eliminate some of the variability in the rate of these reactions where under the exact same environmental conditions with the same castable composition there are seen sometimes differences of more than a month before the castable shows reactions. These tests were expanded by testing samples that were cured and dried at 220oF (104oC) to see if there were any reactions after dry out of the calcium aluminate hydrate phases. This was done to verity if the calcium aluminate phases partially dried out will act as a desiccant and re-hydrate the partially dehydroxylated phases. Technology Breakthrough - Patented Hydro-Lite Coating Removing the alkali containing perlite from the insulating castable composition makes it difficult to manufacture a low density insulating castable suitable for fired heater applications. In addition, the insulating aggregate chemistry is also important where the calcium in the aggregate would potentially form Anorthite which through hydrolysis can be reduced to quartz and kaolinite. These problems were solved by two things: First, by using a calcium free Mullite mineralogy where the Anorthite phase was not formed. This eliminated the Anorthite hydrolysis. Second, to be able to replace perlite's effect in lowering the density by the development of a new patented technology of coating the insulating aggregate with a hydrophobic coating that is chemically bonded to the aggregate. This patented coating covers all the open porosity in the aggregate so that if 4 mesh (6 mm) coating aggregate is crushed to minus 50 mesh (0.3 mm) all the crushed 10 | P a g e aggregate when placed in water will float due to the hydrophobic nature of the coating. This patented hydrophobic coating does repel water and when placed in a castable composition reduces the water demand by 40% to 50%, and, at the same time produces a castable mixture that flows well for casting into forms and a gunning composition that has reduced rebound. This new breakthrough technology enables engineers to develop castables that are not affected by the Hydrolysis or Carbonation if the precautions are taken to choose the high purity cements, clays and other modifying raw materials that do not promote the Hydrolysis or Carbonation reactions. The Hydro-Lite Patent provides a foundation for a new series of castables where: • The soluble alkalis are eliminated • They require 50 % less water to install • It lowers the water to cement ratio to increase resultant castable strength • The gunning installation has been significantly improved with rebound significantly reduced by more than 60%. • The permanent linear change from cast to fired has been reduced by more than 50%. • Faster dryout can be achieved without additional additives which may harm castable consistency. The reduction in permanent linear shrinkage creates a more thermally stable structure for fired heater applications. In summary, the resistance of the composition to AH/Carbonation comes down to a function of purity of the raw materials used as defined by chemistry and by mineralogy. Purity in chemistry relates to the reduction and/or elimination of soluble alkalis that would be found in the clays and cements used. Purity in mineralogy relates to eliminating any raw materials that contain minerals, like albite or Anorthite, that has been known to stimulate and/or promote the AH/Carbonation reaction. Also of the mineralogy the intermediate and lower purity cements form hydrated phases consisting of calcium, silica, magnesium and iron that react and promote AH/Carbonation. 11 | P a g e These aggregate, clay and cement phases, individually or working together, promote the AH/Carbonation reaction. The Hydro-Lite patent opens the door to these new castable breakthroughs and provides various features and benefits not afforded by the conventional castables used in fired heaters. The Alkaline Hydrolysis has been eliminated from the reacting in the composition due to its controlled chemistry and raw materials. Picture 2 presents the results of a three year test in the accelerated AH/Carbonation test of the Blazelite 23 LI AHP composition. During this time there has been no detrimental surface or internal damage to the cubes. Temperatures have ranged from 50oF to 120oF (10 to 49 oC) during this time. Water was added to the pans to replace the evaporated water and pictures taken on a weekly basis. This does prove that these compositions can eliminate the AH/Carbonation reaction and provide a suitable long term replacement for current failing compositions. Picture 2 Comparison of Blazelite 23 LI AHP after 3 Years in Rapid Alkaline Hydrolysis/Carbonation Testing Cast June 24, 2014 Current Picture on June 24, 2017 Technology Applied in Castable Design The use of the hydrophobic coated aggregate has changed the castable composition technology to allow optimized casting and gunning compositions. In addition, it enables the AH/Carbonation proof compositions to be developed using key raw materials. The compositions used in casting or pumping are similar to 12 | P a g e current technology where the formula takes into account the significantly reduced water content. The lower water to cement ratio means higher resultant castable strengths so the quantity of cement can be reduced to maintain the same competitive strengths. In the gunning composition the Hydro-Lite patented coating does provide an option to re-design the matrix that combines good physical properties with an easily gunning composition that has very low rebound. The rebound quantities for horizontal gunning for an experienced nozzle man would be in the 5% to 8% range, not counting cut back. For overhead gunning the rebound ranges from 10% to 17%, not counting cut backs. The casting and gunning formulas with the Patented Hydro-Lite aggregates have a significant reduction in water required for installation which also provides the benefit that a partial dryout as specified by API STD 560 due to control of Alkaline Hydrolysis is not required. However, the conventional 2300 LI castable without the Hydro-Lite aggregate would have more mechanical moisture in the castable after the partial dryout than the Hydro-Lite based 2300 LI castable as cast or gunned into place with no dryout. This significant reduction in water with the density in the 60-65 lbs/ft3 (961 to 1041 kg/m3) can be easily dried out without having any holds at 300oF or 600oF (148 to 315oC). The API STD 936 calls for holds of 6 hours at 300oF and 600oF (148oC and 315oC) for a 6 inch thick lining. These castables can be heated up at 200oF (93oC) per hour from ambient to operational temperature of 1350oF (732oC), and would take just over 6 hours as compared to conventional castables with 2 six hour holds plus heating at rates of 100oF (38oC) that would take a total of over 24 hours. This time savings enables the refinery to get back on line more rapidly. If another castable installed in the refinery not using the Hydro-Lite aggregate technology is placed in the lining then the refinery must be brought up to operational temperature at the rate that of the conventional castable can withstand. 13 | P a g e Risk of Failure Due to Hydrolysis and Carbonation - Repair of Convection Sections The purest clays, cements and other raw materials are used in the manufacture of these AH/Carbonation proof compositions. The higher the purity, the higher the cost of the castable. Yet its design is such that the AH/Carbonation reaction is controlled and long term storage will not cause/promote the reaction because the primary reaction drivers have been eliminated through selective choice of high purity raw materials. The potential repair cost of AH/Carbonation reacting in a convection section is very high due to the need to remove the special stainless steel finned tubes to repair the lining. There are several sections of work that are required to repair the convection section. The cutting of the tubes and removal from the convection section. Secondly the removal of the castable and anchor replacement/repair prior to new castable being installed. Then replacement of the castable into the convection section and finally the specialty welding required to bring the specialty stainless steel tube back to specification to meet the application pressure and temperature. The repair of one convection section could cost into the millions of dollars. The use of these Hydro-Lite castables, like Blazelite 23 LI AHP, mitigates the cost of repair and eliminate the man-hours required to monitor and maintain these fired heater modules prior to refinery start-up. Summary Our solutions significantly improve current engineering, design and function of current fired heater insulating refractories. However, the rapid deterioration of insulating castable linings stored longer than 4 weeks is one of those problems where the project experience continues to prove that the risk of Alkaline Hydrolysis/Carbonation is significant to the Owner with or without partial dryout. There is a serious gap in the support provided by the supplying vendors and manufacturers to avoid these problems. This leaves the purchaser without a dependable cost-effective alternative or a reasonable level of accountability for satisfactory results that is desirable from a project execution basis. However, significant innovation in the engineering of insulating castables used in Fired Heaters have created compositions using patented hydrophobic insulating aggregates and high purity raw materials designed to control and eliminate Alkaline Hydrolysis/Carbonation and the deterioration of the lining compositions. 14 | P a g e These proprietary compositions, using the patented hydrophobic aggregates, create a compositional framework where optimum casting and gunning properties can be designed. This is demonstrated in the gunning compositions with very low rebound and ease of installation. These compositions not only eliminate Alkaline Hydrolysis and Carbonation reactions, they also reduce the water demand for installation by 50%, maintain the density and strengths required by the application, enable the composition to not need a partial dryout, but can be rapidly dried out after erection and during start-up of the refinery operations related to the heater. This, along with creating a more thermally stable structure that has 50% less shrinkage and higher strengths at temperature, provides the Owner with a product that has accountability with a significant cost benefit by the elimination of AH/Carbonation reactions, elimination of partial dryout costs, and enabling rapid firing during initial start-up of the refinery. Acknowledgement We thank Robert L. Antram, President - Antram Refractory Consulting, LLC and John R Peterson, President of Refractory Consulting, LLC for their technical insight and support providing refractory application information on fired heaters for this paper. Appendix 1 15 | P a g e Section A Calcium Aluminate Hydration Reactions Calcium aluminate cements when mixed with water chemically react with the water and forms calcium aluminate hydrates. Immediately after water is added the calcium aluminate going into solution becoming Ca+ and Al+ ions. This reaction can be shown in the following equation. CA(s) + water CA(sh) + water CA = calcium aluminate (s) = solid Ca(H20)2+(l) + 2Al(OH)4- (l) (sh) = surface hydroxylated (l) = liquid This shows a reaction first on the surface of the cement which dissolves the Ca and Al as ions in liquid solution. The concentration of ions in solution increases to a saturation point, at which time they precipitate as structured hydrates. As the structured hydrates are precipitated out, the solution becomes unsaturated, allowing more Ca2+ and Al3+ ions to go into solution. This creates a cyclical pattern: ions coming out in the form of hydrates…ions dissolving from the CA to supply more ions to saturate the solution, then precipitation, then dissolution and so forth until there are no more ions to donate to solution and all the CA is consumed in the reaction and all the usable ions are precipitated as hydrates. The primary hydrates formed in calcium aluminate hydrates are listed below as a function when formed at the various ambient temperatures. Temperature of Curing o o C F <10 <50 Hydration Products CAH10 + AHx (gel) 10-27 50-80 CAH10 + AHx (gel) C2AH8 + AH3 (gel) >27 >80 C3AH6 + AH3 (crystal) Conversion Reaction >27 > 80 Courtesy of Kerneos CAH10 > C2AH8 + AH3 (gel) C2AH8 >C3AH10 + AH3 (gel) Appendix 1 16 | P a g e Section B Carbonation Reactions Diffusion and reaction of sodium ions in castable matrix1 2Na+ + 2OH- Na2O + H2O 2Na+ + 2OH- + CO2 + H2O Na2CO3 . xH2O Carbonates then react with Calcium Aluminate Hydrates forming calcium carbonate and sodium aluminate hydrates. Na2CO3 + CAH10 CaCO3 + Na2O.Al2O3 + 10H2O Sodium carbonate is then regenerated by additional CO2 reacting with the sodium aluminate. CO2 + Na2O.Al2O3 Na2CO3 + Al2O3 From the initial reaction of Na+ and CO32- the CA hydrate is consumed and replace by calcium carbonate and alumina gel. The calcium carbonate is found in forms of aragonite, vaterite and calcite.9 The alumina gel is found in forms of bayerite, nordstrandite and gibbsite. These phases can be identified by X-ray diffraction methods. 2Na+ + 2OH- Na2O + H2O 2H+ + CO32- CO2 + H2O 2Na+ + CO32- + CaO.Al2O3.10H2O CaCO3 + 2Na+ + 2Al(OH4)- + 8H2O 2Na+ + 2Al(OH4)- + CO2 + 3 H2O Overall Reaction 2Na+ + CO32- + 2Al(OH)3 + 7 H2O CO2 + CaO.Al2O3.10H2O Courtesy of Kerneos CaCO3 + 2Al(OH)3 + 7 H2O Appendix 1 17 | P a g e Section C Albite and Anorthite Hydrolysis Reactions Examples of Hydrolysis of Albite and Anorthite. Albite Hydrolysis 2NaAlSi3O8 Albite + 2CO32- + 3 H2O Carbonate Ions Water Al2Si2O5(OH)4 + 2Na+ + 2HCO3Kaolinite + Sodium Ions Bicarbonate 4H4SiO2 Quartz Anorthite Hydrolysis CaAl2Si2O8 Anorthite + 2CO32- + Carbonate Ions 3 H2O Water Al2Si2O5(OH)4 + Ca2+ Kaolinite + Calcium Ions 2HCO3Bicarbonate 18 | P a g e Bibliography and References 1. C. Parr, C. Alt, J. Pelletier "Carbonation Caused By Alkalis", Technical Sales Presentation, Kerneos, Inc, 2011 2. C. Parr, J. Pelletier, "Hydration Schematic for CAC", Email Communication, pp 1-2, (2013) 3. W.H. Gitzen and L.D. Hart, "Explosive Spalling of Refractory Castables Bonded with Calcium Aluminate Cement", American Ceramic Society Bulletin, Vol. 63, No. 7, pp 905-910 (1961) 4. G. MacZura, L.D. Hart, R.P. Heilich, and J.E. Kopanda, "Refractory Cements", Ceramic Proceedings, Columbus, Ohio, 1983. 5. J.E. Kopanda, G. MacZura, "Production Processes, Properties, and Applications for Calcium Aluminate Cements", Alumina Chemicals Science and Technology Handbook, Section III, pp 171-183 6. I. Pundiene, S. Goberis, V. Antonovic, R. Stonys, A. Spokauskas, "Carbonation of Alumina Cement-bonded Conventional Refractory Castable in Fireplace", Materials Engineering 2006, Kauna,Lithuania, 2006 7. G.V Givan, L.D. Hart, R.P. Heilich, G. MacZura, "Curing and Firing High Purity Calcium Aluminate-Bonded Tabular Alumina Castables", The American Ceramic Bulletin, Vol. 54, No. 8 (1975) 8. D.B. Ellson, W.W. Wright, "Self-Destruction of Unfired Refractory Castables", UNITCR Congress, pp 475-486, 1993 9. L.D. Hogue, W.A. Jackson, "Nature of Carbonation of Hydrated Calcium Aluminate Cements in Castable Refractories", Industrial Heating, August 1997, pp 45-49 10. S. Sakamoto, E. Kudo, "Carbonation of Alumina Cement-Bonded Castable Refractories", Journal of the Technical Association of Refractories, 20[1] 18-23, (2000), Japan 11. Andrew McLeish, "Geological Science", pp 67-99 Nelson Publishers 12. Francios M.M. Morel, Janet G. Hering, "Principles and Applications of Aquatic Chemistry", John Wiley and Sons, pp 274-290, 1993 19 | P a g e |
ARK | ark:/87278/s6pk4s4d |
Setname | uu_afrc |
ID | 1388793 |
Reference URL | https://collections.lib.utah.edu/ark:/87278/s6pk4s4d |