Refractories in Reforming Waste Heat Boilers.

Refractories in Reforming Waste Heat Boilers.

© Pieter Barnie Enslin.

This paper focuses on refractories in (WHB’s) Waste Heat Boilers, specifically in the Petrochemical Reforming (ATR) Auto Thermal Reformers, and (GHHER) Gas Heated Heat Exchange Reformers. SMR’s (Steam Methane … ) and to some extent SRU’s (Sulphur Recovery … ) and other related units may also apply.

The late and honestly great Dr. J.D Hancock wrote in his book: Practical Refractories; refractories are like “Black Magic”. This statement combined with some unique experiences led me totreat refractory applications as though it has its own unique “personality”. It allows me to engage with an open mind, as opposed to stop, and make or come to a conclusive decision when one finds some supposed generic failure or occurrence. “… oh it failed because of this or that and that’s it….” No, this is not always the case.

It’s typically that there could be more to a failed refractory than meets the eye. If we miss these “hidden” elements, then we miss an opportunity for progressive improvement. Only a few failure-, damage- & attack mechanisms are generic, and even then, they typically occur due to, or for very specific reason. Often failure is only unravelled after in-depth review, considering varying process conditions, upsets, refractory material qualities, workmanship, and a number of variables I address going forward.

“The less you see the more you believe (know)” This approach is dangerous because it could mean we ignored valuable history, trends, or information from various sources. Quote: Jacques Tourneur – Horror movie director.

But on the other hand. After a few investigation sessions I noticed that – “The more you see the less you know”. Yes, eventually a fair outcome is reached from the available information, but there comes a point in some inspection assessments that leaves me in wonder. I think this is where Refractories and “Black magic” meet. Yes, I quoted the Red-Hot Chilli Peppers.

Fact is there are always some elements or attributes involved that affect refractories and their ancillaries & auxiliaries uniquely. One should best not assume a result or cause without careful analysis of all information. It could pass as ignorance which could cost millions!

To conclude this section and move ahead while in a manner repeating myself: When focusing on refractories as a solution one should consider all aspects: From design to process, overall history, workmanship, and everything that has gone into the manufacturing, placement, and management of the correct lining. The one thing I want to add to this holistic approach is – the client. It’s not about anyone other than the client and the integrity of their refractory lined equipment. As a specialist I am are required to add value – I trust I am doing this to the best of my ability and believe herewith to encapsulate some principles in a condensed manner. In addition, I trust this may offer some guidance during the decision-making phases for the next refractory challenge you may face.

Content

Topic	Page Number
Introduction	3
Equipment	3
Process	4
Types of refractories, linings, and attributes	5
Installation & methods
Demolition and shell preparation	9
5.2 Anchorage	9
5.3 Monolithic (unshaped)
5.3.1 Gunning	11
5.3.2 Casting	12
5.4 Brick	15
5.5 Tube sheet & ferrules	17
5.6 Manway	19
Expansion	19
Dry out	19
Types of failures
8.1 General	21
8.2 non-process	22
8.3 Process conditions
8.3.1 Fouling	23
8.3.2 Metal dusting	23
8.3.3 Carbon	24
8.3.4 Silica leaching	25
8.4 Other	25
Maintenance issues	26
Research and development	27
References	28

1. Introduction:

It has always been my intention to help my clients with the most accurate degree of information and not let them have blind faith in what I do. To do this I rely on my sense of curiosity, my knowledge, my experience, but also the knowledge and experiences I had with peers, given I am in the fortunate position to be working with some of the best minds and hands in the industry. Today I am a freelance inspector and for some projects an inspection supervisor -which implies full time onsite oversight – ensuring projects run as smooth as possible, without compromising quality, and adhering to schedule. I have also given some basic refractory training, but we will see what the future holds in this regard.

I learned not to take uneducated & unwarranted “chances”, and not to run trials at someone’s expense. Without substantial proof that a recommendation will work for as long as it needs to within certain parameters one might be taking unwarranted chances. It’s important not to run with hypothetical assumptions, but instead actual relatable proof to support that which one may be doing to improve a situation, or to manage risks. As result, this is my attempt given my current level of knowledge of a factual summary – besides for a few hypothetical and falsifiable scenarios that I will point out. 

I will also offer an overview illustrating that a general understanding of the equipment, its operation, the mechanical & chemical mechanisms, as well as an understanding of applicator and material constraints can helps one make the most informed decisions.

Among some of the points considered are materials, the various processes, and my favourite; man vs machine, or if you like; workmanship vs operation. I am referring to ops. in the widest possible scope considering installations and unit operation. 

I trust it will add value to anyone aiming to learn or perceive the subject from a quality improvement point of view.

2. The Equipment:

WHB’s or heat recovery steam generators are widely used in processing industries to improve energy efficiency.

Essential components required for waste heat recovery are:

– A source of reliable and consistent heat.

– A recovery system to generate steam.

– End-use for the recovered (heat) steam for preheating, electricity supply, or process control in other areas of the plant.

Thanks to the unconventional energy source of WHB’s it has unique design features. This design is typically backed by high alumina refractories of which goals are to contain the heat, insulate the unit, contain the gas product, and optimize or lower maintenance cost ensuring minimal downtime and production loss. 

In reforming the most common WHB design is a cylindrical steel shell with an interior of two opposing metal sheets each connected with a collection of horizontal tubes forming the boiler section which is filled to a specific level with the water required.

The cylindrical drum is divided into segments: manway or inspection door, dish end, barrel, the reforming to barrel interface or throat section, and the tube sheet.

3. Process:

The organic feedstock (e.g., natural gas) and steam (and sometimes carbon dioxide) are mixed directly with oxygen and air in the reformer. The reformer itself comprises of a refractory-lined vessel that contains the catalyst, together with an injector located at the top of the vessel. Partial oxidation reactions occur in a region of the reactor referred to as the combustion zone. It is the mixture from this zone that then flows through a catalyst bed where the actual reforming reactions occur. The heat generated in the combustion zone from partial oxidation reactions is utilized in the reforming zone so that in an ideal case the process can achieve thermal equilibrium.

Product gasses are then introduced into WHB (Figure 01 Configuration) system at around 750-1050°C from the ATR through a diffuser (Mushroom) & transfer line. In most ATR designs, heat is quenched with steam at the inlet because it exceeds safe gaseous temperatures with enhanced corrosive abilities concerning the boiler tube and other metallic surfaces eg: bypass valve brackets etc… In the event it flows to a connected GHHER the quench function is sometimes bypassed for sufficient energy for the endothermic reaction.

…

The hot equilibrated gaseous product enters through the hot tube sheet side during which it creates steam. Heat is transferred from one medium to another when natural water circulation occurs in the boiler with the differences in density of water heating up. The gas product exits at the cold side thanks to a pressure differential or mechanical draft @ approximately 350 to 480°C where it is directed for Syngas processing further downstream. A WHB in GHHER configurations is generally significantly smaller due to lower cooling duties required and limited steam which can be created. These WHB inlet temperatures are around 560 to 640°C.  

In many designs, the tube sheet also consists of a bypass valve used to advance or retard flow in the event cold side temperatures rise to detrimental levels and vice versa.

A WHB’s practicality is measured by its steam to fuel efficiency: The ratio of the heat added to the boiler feedwater to produce the output steam to the amount of energy inputted with fuel.

Generally, the consumption of waste heat is considered a definite benefit as it recycles waste, is good for the environment & business both financially and pragmatically. It is however a given that for the by-products of upstream processes it is not the most efficient fuel source and WHB’s are less efficient than other types of boilers.

Steam is used for the transportation and provision of energy.

The benefits are that it is efficient and economically sound to generate, easy to distribute, easy to control, easily transferred to the processes and a steam plant is relatively easy to manage while the process is flexible. Process and in-service checks involve temperatures, pressures, boiler feed water quality & consumption, flow rates, hydrostatic testing, and ultrasonic testing.

Fig 1: Typical arrangement (image borrowed)

To conclude with relation to the process, it is important to note that the relationship of steam to the pressure at which it is generated could be fairly simple in principle. But what is not as simple is to strike the perfect balance between pressure and temperature all in providing the best energy solution. These units operate under pressure but the higher the pressure of a boiler is the more heat must be applied to make steam. (Note the segment in Dry outs – Figure 11). High pressure – high temperature steam contains more energy per unit, which is known as Enthalpy.

Enthalpy: The thermodynamic potential made up of energy and pressure. More heat and pressure do not necessarily mean more energy, it is possible to get more steam per ratio at a slightly lower temperature than simply assuming more heat equals more steam energy.

 4. Types of Refractories and Linings

The performance of refractories in a WHB are measured by the chemical, physical, mechanical, mineralogical, and thermal properties which all should be compatible with the application.

In further considering material types or expected quality characteristics one should look at process parameters, temperature profiles, mode of operation, chemical environment, operating pressures, atmospheres, quality of refractory, and workmanship.

Generally, the linings in reforming WHB’s are dual or three-layer linings (See Figure 02 Photo of multiple/three-layer lining), I have however come across single-layer linings as well. Lining thicknesses may vary from 150mm to 340mm in total. Factors such as water-cooled jackets could also affect the lining configuration.

Figure 02 internal
Hot face – High Density Brick              Intermediate – Bubble Alumina Brick      Cold face – cast Note the oil between each layer

A generalized description of common linings:

1) The barrel, transition, and dish end. 

a) A monolithic insulation backing lining with insulation brick and high-density brick to conclude the three-layer lining 

b) Insulation and hot face both monolithic

c) Insulation monolithic and high-density brick.

2) The tube sheet: 

a) Single high alumina castable.

b) Dual layer bubble alumina backing and high-density high alumina front face.

3) The manway is predominantly a combination of pre-cast shapes and wool placed in a monolithic calcium aluminate section. Or it could be monolithic installation either gunned or cast, the orientation of the manway dependant.

4) The cold side is generally made up of a medium weight insulating calcium aluminate grade of refractory.

Considering all the required criteria with a WHB the type of refractory applicable on the hot or process side is the 99% Alumina’s. This face of the lining is exposed to the most severe conditions, and it is critical to engineer it correctly. Therefore, white corundum or other Synthetic Alumina (Tabular & White Fused) is applied. The chemical composition, method of manufacturing, and refractoriness are the best suited for this application.

There is no silica in these hot face section refractories, and for best practices neither in the backing layer or wool/fibre product implicated either. Silicas could leach from the lining and cause sensitive downstream damages. The hydrogen environment with related pressures is not conducive to stable Silica. I will elaborate on this in section 8.6 Silica leaching.

The high purity Alumina can also easily be used in shaped or monolithic form. It is the “more pure” forms of alumina which are stable in alkaline environments & have dimensional & volume stability, hardness, high-temperature mechanical resistance or if you like thermo-dynamical stability as well as exceptionally good chemical inertness. 

White Fused Alumina is calcined to very high purity. The end product as per the name is produced through a fusion process. Tabular alumina is sintered calcined alumina. However, these have much in common the sintering process once cooled correctly allows for a specific uniform microstructure that brings about low residual porosity & very high densities. It also has a unique texture or coarseness which promotes a good matrix bond. These products are calcined at temperatures of as high as 1250°C+ for extended periods dependant on the desired crystal or microstructural formation pre-fusion or sintering.

The hot face brick or high duty alumina bricks as explained above are designed for load bearing, they have good volume stability at high temperatures because they are dry pressed or fused (no water content) to the desired shapes with a tongue and groove design which assist in structural stability and prevention of gas bypass. Average thermal conductive values are high at around 2.7 to 3.8 W.m-1k-1 at temperatures around 1000°C.

The following graph (fig 03) illustrates alumina with higher levels of “impurities” indicating that the higher the hot face lining alumina level the higher the densities & lower porosity. Note the lower Sio2 levels which are required for the environment.

   Figure 03 Alumina Purity

Insulation bubble alumina bricks have a thermal conductivity of around 0.20 to 0.90 W.m-1k-1 and a very low Mpa usually below 3. They are thus not load-bearing and are typically designed with a narrow hot-face side (less of a structural bond than the hot face bricks) to turn a radius, no importance is taken away from these bricks as they are critical in assisting the temperature drop curve and insulating the unit.

The mortar must be a grade higher than the brick which it bond, besides the bond function it must also be able to sufficiently seal spaces between brick and compensate for dimensional variances in the lining – note that this is not a desired function in pressurized hydrogen environments! Brick inspections must validate that the already tight tolerances are met, warpage or imperfections in bricks or pre-shaped forms should not be a concern when building.

Phosphate bonded mortars could be used as it offers improved volume stability with improved resistance to vapour attack. Just make sure that your (P205) Phosphate pentoxide content is sufficiently high or else it will fail! When requesting a high-grade alumina Phosphate bonded mortar aim for 4% phosphate content. I have experienced critical lining failure due to inadequate addition of P205, the mortar disintegrates to allow for a total lining collapse.

Monolithic – dense and insulating – are used for their ease and speed of installation plus because they eliminate joints with inherent weaknesses.

High purity calcium aluminate cement (The combustion mixture of limestone and bauxite -containing a minimum of 70% Al2O3) is used for early strength development. It imparts good hydraulic setting properties when mixed with potable water ph 6-8, Chloride ≤50ppm % Temperature @ 18-28°C. Most specifications have a wider temperature scope. Note: Phosphate bonded materials are also applied in certain cases

During hydration the hydrate is heated thanks to an exothermic chemical reaction. Water molecules break free of the complexes they have formed with the ions in the crystal lattice. The loss of the water molecules then changes the structure of these complexities hence the properties which initiates the initial set.

Later during heat-up, (discussed in more detail during the dry out phases) transformation occurs by further volatilizing of liquid to aid the formation of a complete hydraulic bond which with increased temperature converts to a ceramic bond.

Other attributes:

Thermal conductivity indicates the general heat flow characteristics of a refractory. It is dependent on factors such as its chemical & mineralogical composition, apparent & total porosity, permeability, the extent of sintering or fusion as well as pressure and temperature in service. 

Refractories with lower thermal conductivity are preferred as insulation against the shell side as they assist in conserving energy. High-density high purity alumina refractories tend to have an inverse thermal conductive behaviour meaning that the higher the temperature raises the lower its thermal conductive value tends to become. A 20% variance from published values is not uncommon when in practice values are compared at higher temperatures. Hydrogen atmospheres also increase values from test values as tests are normally done in standard atmospheric environments.

Thermal expansion occurs at elevated temperatures and consequently, dimensions change and compressive loads increase. Note: A shape /brick tested in a lab expand equally, and a shape exposed to process expands irregular – the process side moves different to the back or sides of a brick… the irregular movement creates progressive deterioration noted as surface cracks or/& pinch/mechanical/thermal spalling.

Reversible thermal expansion values for High Alumina refractories are between 0.6 to 0.9% meaning that shapes expand and contract back to near original size.  

5. Installation methods

5.1 Demolition and shell preparation:

Normally installation will be preceded by demolition of previous linings which would require one to ensure a clean shell free from surface rust, solvents, and any deviations such as excessively deep marks made with moils when using jackhammers during the demolition stage. Shell preparation or cleaning can be done according to clients’ specification requirements typically guided as per Figure 04. The API 936 specification could also assist with guidelines.

Figure 04. Shell prep guide.

Note: It is critical not to damage the tube sheet, welds could be compromised among others, and repairs will set schedules back plus cause un-planned breakdown repairs. Personally, in understanding this I have implemented demolition training and Standard operating procedures for demolition teams to incorporate in their method statements and quality control plans.

5.2 Anchorage: 

Consideration for the correct choice of anchors: 

Metallic grade, length, form, diameter, and means of securing the anchor to the shell. Anchor placement, pattern, distance apart or density, wax or bitumen coating, and/or plastic cap addition. Given differential expansion ratios in relation to the monolithic which it is designed to restrain. 

When considering a metallic grade for an anchor it is best to assume the process or hot face gas temperature instead of the temperature to reach the insulated anchor. Grades are generally 310 or 316 stainless steel as it is commonly available and offer stability in the given environment. These composites do not easily degrade under these types of temperatures, overall conditions, or with the chemical compounds within the monolithic which it stabilizes. Should the refractory hermetic seal be broken they also offer resistance to the process gas, excelled temperatures, and carbon deposits.

304 stainless steel anchors do not react well in oxidizing environments cycling near 1000°C, even if it is protected by the insulation around it the risk still exists. It also acts poorly in carburizing environments and lining retention failure may lead to critical failure.

The general conception is that 304 stainless steel anchors may apply to the cold side of a WHB however it is more common than not that 310 stainless are used throughout even though the “environment specific” justifies said grade.

That said I have experienced scenarios where Inconel anchorage was the standard. It is the best in terms of resistance but also the costliest.

Using the incorrect metallic grade as a restraint system could introduce failure mechanisms such as among other sigma phasing. 300 and certain 400 grade stainless steels are susceptible. It is brittle fractures in the metallic structure because of exposure to high temperatures that “leads to” restraint failure. It seems to occur in the 530°C to 930°C range and Embrittlement can result by holding within or cooling through the transformation range. The question could then be why it doesn’t fail in process and only as I stated – leads to – failure? It seems that the attack occurs in process and actual fracturing occurs at less than 260°C.

There could also be instances where carburization may occur if the substrate is exposed to temperatures in the region of 600deg C. Carbon deposits can increase thermal conductivity in a lining but it can attack the ferric element in certain metallic grades content dependant. Volume increase is the most evident sign of severe carburization.

The length of an anchor should be 65 to 75% of the thickness of the monolithic lining. The diameter would typically be 5.5mm, 6mm, 8mm or in instances upto10mm. The diameter and shape are determined by the mechanical restrain requirements in part dictated by the mass of the monolith which they need to restrain. They could either be hand or stud welded or bolted into position. Smaller tube sheets often have no anchorage other than the ferules which inadvertently however debatable offers a degree of restraint to the lining – more specifically metallic ferules.

It’s worth remembering that the tube sheet cast is either cast- or rammed in behind the adjacent brick lining aiding additional prevention from detachment and preventing gas bypass on a straight joint. Note that hex ferules are only secured by the lateral force and a rammed periphery – there are no anchors either – they tend not to “walk” such as bricks do.

The inclusion of anchors on the cylindrical section of the WHB is debatable because it often is packed in behind a self-supporting brick arch. I have experienced the absolute minimum number of anchors being installed waiving any specification in terms of placement and spacing. No resulting failures have been recorded where I was involved

The anchored pattern should be staggered (See Figure 05) with anchor tips not overlapping or getting closer than 25mm (± 1inch) to each other. This can create the tendency for cracks to grow toward each other given anchors can tend to cause micro hairline cracks into the refractory lining if inadequate expansion allowance was given. For clarity: metals and ceramics do not expand at the same ratios. So, when caps and/or full coverage coatings are not applied for expansion, it could cause premature breakages. If cracks are formed from the same direction they could meet up and may become panelled sections of severe cracking with thermal cycling. 

Figure 05 Typical anchors patterns

Some clients are happy with zero expansion allowance with anchorage given that they have not experienced the resulting failures in the past. Or it could just be that their maintenance strategies do not allow time sufficient for deterioration to reach detrimental levels. Plus, there is the understanding that almost all monolithic materials have a negative (PLC) permanent linear change value. This implies that material shrinkage and the additional porosity created during dry outs could be sufficient for anchor movement. This naturally applies more to insulating material. However, physics dictate that the dissimilar materials (metallic anchors and ceramics) do not expand at the same ratios. Both expand more at the hot face side but anchors even more so. Even though anchor expansion is manageable by placing plastic caps over the anchor ends, this compensates for the area’s specific expansion closest to the hot face temperature but not as a whole for the total area of the anchor. Thus, it’s typically recommended that in dense linings they be coated with bitumen or wax that can burn away without residual ash and leave a total gap around it.  

5.3 Monolithic linings.

• • Gunning:

With gunning achieving the right or consistent densities becomes problematic because it can run higher in areas if pressures are too high, or if plant pressure is inconsistent. An adequately sized mobile compressor capacity 5 – 7 bar is sufficient. Air must be filtered and clean without any traces of oil or contamination – use a water trap.  Maintain and keep air pressure steady at minimum optimal value to prevent loss of binder and to reduce rebound but still ensure sufficient compaction.

Water supply must be clean and not run past heated tubes steam lines. Pressure must also be adequate with 10 bar (or at least more than the air pressure) ensuring sufficient mixing and hydration. Plants run at much lower pressures, so use a water pump. Controlling water at the nozzle, especially with low cement material requires competent operators and artisans. Maintenance parts must be replaced at every given opportunity. I will elaborate on prequalification below in the cast section.

One would typically only gun in large vessels, not directly overhead nor below 30 degrees to horizontal. Should the decision be made to gun then rebound entrapment, laminations, and inclusions of cavities must be eliminated. Ensure swift sweeping or circular nozzle movements to prevent entrapping rebound, Rebound is expended or improperly hydrated material which bounces off the surface. In a sense of speaking, I can say that one should install around whatsoever material deflects.

If rebound collects in corners and anchors, it can be easily removed by shutting off the material at the nozzle (if this function is available or ask the operator) and turning up the air pressure air to blow out the corner or area where the rebound has collected. Alternatively, the cutback assistant must get to it timeously.

Personally, shotcrete (wet as opposed to dry gunning) has not worked well where I was involved at a few projects but people that I know has enjoyed great success with it. The exciting part is that technology and skills are advancing, and I am looking forward to seeing where we will be very shortly. It’s traditionally marketed as a cleaner, less wasteful, and a faster method hence I am convinced mind sets and ease around this method of installation will adapt soon.

As I am moving over to Casting, I must say that am not taking anything away from gunning, it should not be considered inferior – there is a time and a place to apply it. The principles of a homogenously installed inclusion free, sound lining applies across the board, but is typically easier achieved with casting. Even though gunning is not always practical in some aspects of a WHB it is still used by specialist teams where technical restrictions for nozzle men can be overcome.

• • Casting:

Besides being comparatively time-consuming for large area installations casting has definite advantages over gunning specifically in a narrow horizontal cylindrical application.

It is important to ensure a sufficiently sized and type of mixer able to shear as close to the pan sides, floor, and shaft as possible. Backup mixers (and gunning machines) are always recommended but seldom in place until it is too late. Five to seven minutes are considered sufficient time for ensuring a good mix in a standard pan mixer; however it varies based on the chemistry in question as 2 to 3 minutes applies to certain material.

Pre-installation tests (this applies to gunning as well) of the equipment, operators and specifically the material should be done to assess workability, product quality, and labour competencies. A set means of mixing and application must be agreed upon before the installation may start, typically a material supplier should supply an installation instruction, these tend to be very conservative.

Mixing the materials should be done as per the installation instruction or if not available the datasheet, I prefer the batch test certificate for certain non-generic values such as batch specific water addition. If pre-installation tests contradict documentation a concession and/or on-site attendance by the material supplier is required. Overall, this process allows one to assess the materials workability, setting time and general ease of installation attributes and has the potential to spare frustrating blocks when one starts the installation.

Best practice dictates that samples be made and test before installation because test certificates are constructed under lab conditions where the environmental elements and available resources like water may differ dramatically.

All protrusions, nozzles, and pipes into the vessel are typically wrapped with 2-6mm ceramic fibre paper. Waterproofing with plastic tape or membrane is mandatory to prevent water absorption & dehydration of the immediate lining. It can induce early failure in the form of embrittlement cracking. Here is a guideline if drawing does not give detail about wrapping thicknesses: 

• • 50 mm Ø < Nozzle < 125 mm Ø: 2 mm

• • 125 mm Ø < Nozzle < 300 mm Ø: 3 mm

• • 300 mm Ø < Nozzle < 600 mm Ø: 6 mm

The shuttering system whether metallic or in the very specific case of wood must be lightly greased or oiled (mixing a little grease and oil makes for a nice consistency). It should be able to contain the wet refractory mass. A tight fit between the shutter and the shell as well as between lift panels must be ensured. 

Lifts or panels should be managed in workable sizes to ensure that the best possible installation may take place in other words ensuring no cavities or voids are included in the placement process. To accommodate high-frequency vibration and fit the pokers sufficiently the shutter lift heights are best kept to 400mm or the length of the poker. It helps to make sure the anchor pattern allows poker access. It also helps to keep the poker cool by having a pale of water nearby to store it in when rotating pokers as it tends to warm up within the lining. The combination of an aggressive exothermic reaction and mechanical actions all ads as heat sources potentially limiting workability time.

Special attention must be given not to exceed the usability period of the mixed material because it will set off and cause laminations in the lining. Proper and consistent mixing times should be ensured, and the mixer and buckets must constantly be kept clean from older material. The process of keeping them clean involves water, so excess water must be removed properly as a wet bucket can impact material characteristics causing an artisan to think an operator added to much water. 

Insulation castable should essentially not be vibrated, if required to move it into place vibration should be kept to a bare minimum, over vibrating will remove the insulating air and densify it. It can also cause segregation of the aggregates and matrix bonding materials allowing for inaccurate heat transfer and insulation values. If immersion vibrators are used care must be taken to avoid cavities in the lining. 

Dogleg or tongue & groove designs are recommended for the stop-ends given the pressurized gaseous environment and the added measure for by-pass prevention on a joint. I dislike 90 degrees doglegs, a 45-degree edge or a tongue and groove design has less of a tendency to break off at the squared edge’s inherent shear point. 

I have noticed instances where construction joints were kept to the bare minimum to allow the lining to “find its own set of crack planes” – very specifically in the 3rd layer behind two offset brick linings. Inspecting these areas – as time allowed – I could not find or identify any real compromising signs of cross-sectional ingress or mechanical compromise related issues, but good practice dictates that cast panels are typically kept within 1.2 -1.5m². Among some reasons the sizing aids material workability during placement in the given area, thus preventing gelation and as a result horizontal or vertical laminations and poor homogenous bonding. It is worth noting that one pan mixer takes time to mix & dispense, and this must be considered in line with environmental conditions.

If you can – get the material out in one dump. The continued mixing adds energy and in certain material promotes “wetting out” also affecting material performance.

For continuous gunning, larger panels may suffice thickness and material workability dependent, but for casting one should consider the time it takes to mix and place a batch. Vibration also considered before the material starts setting off, as a result introducing the risk of a poor bond if it shifts in a semi-set position, ultimately shearing apart.

Uneven surfaces between cast and brick linings must be ground flush, the design diameter must be maintained, and no warpage (or very strict limits) is allowed, deviations in this area may not be corrected with mortar or levelling powder. Once the monolithic lining is complete a layer of oil paper must be applied to achieve a slip joint between materials should it be a second monolithic lining or brick considering that mortars may fuse adjacently. 

Movement in a vessel occurs throughout but special attention at the interface between the tube sheet and vertical transfer line / inlet / gooseneck / T-section is important. Any obstruction will cause irregular cycling – movement and poorly settled linings or even breakage. As a result the expansions must be perfectly built and correctly offset between layers of the lining, with the correct grade (density, chemistry and temperature rating) ceramic fibre paper. If blanket or bulk fibre is used it must be compressed to the required thickness.

Samples must be cast as per a pre-agreed method, usually, such standards would conform to API 936 if not the client’s spec. stating a set consisting of 4 samples per & shift per batch for the required physical and less common chemical testing. Typically, the tests are for Compressive strength as guided by the ASTM C133 standard. Verifying densities to note compliance with spec sheets. Permanent linear change following ASTM C113 verifying volume stability after a single fired session complies with data sheet values.

Abrasion resistance as per ASTM C704 if and when the material is exposed to abrasive or high velocity areas, not relevant to a backing layer. Chemical tests to verify the material structure and material qualities down to Loss on ignition and trace elements can also be carried out to confirm that the material complies in all aspects. Tests such as Modules of rupture, thermal shock resistance (in quench areas), apparent porosity and thermal conductivity can also be carried out if required. Remember that we are not talking about the ATR here and target tiles related to thermal shock tests is a different conversation.

• • Brick:  

The same principles apply whether it is insulation or high-density hot face brick being installed. Dimensional and damage tolerances would depend on the specification at hand however a guide as per fig’s 06 & 07 is sufficient for most applications – not necessarily relevant to Certain gaseous environments. Additionally, tolerances in the taper shall be ± 1,0mm of the nominal taper where applicable. Parallel cracks over the face of brick are unacceptable unless they are hairline. Cracks over 25 percent of the surface length on insulation cannot be allowed hot face brick cracked more than 5 percent of the length on any face should be avoided.

 Figure 06 

(Guideline not relevant to all environments)

Brick Dimensions (mm)	Dimensional Tolerance (mm)	Warpage Tolerance (mm)	Out of Squareness Tolerance (mm)
≤ 150	± 1,0	0,75	1,5
> 150 ≤ 200	± 1,5	1,0	2,0
> 200 ≤ 250	± 1,5	1,25	2,0
> 250 ≤ 300	± 2,0	1,5	2,0
> 300	± 2,0	1,5	2,0

Voids or chips on the surface are acceptable if less than 10 mm in diameter and 5 mm deep, and no more than 3 per cut surface on insulation brick and half the above figure on hot face brick. Warpage in bricks is not acceptable even though mortar and ceramic fibre paper may compensate for such deviations it is best to avoid it altogether considering hydrogen/methane gas content and potential bypass. ASTM C 134 is a good guideline for physical brick parameter checks. In cycled linings, 2mm Ceramic fibre paper mortared in or mortars alone are commonly used to close cracks or openings in hot face brickwork. 

“Bond” in brickwork is an arrangement by which the joints between bricks in rings or rows are arranged. Figures 08 & 09. Joints should not meet to form straight joints instead they should be offset to ensure better stability in construction as well as airtightness. Ultimately it removes the risk of corner-to-corner mechanical pressure pinch spalling as well.

Brick rings and headers must be joined and sealed with mortar, mortar joints must be kept at <2mm installed thickness, and window framing must be avoided. This is a phenomenon where mortar is applied to allow for air entrapment between bricks.

Closing and keying of bricks are very important to ensure build stability. No bricks smaller than 50% of the original size may be used and no two cut bricks may be installed next to each other. (check your spec. some may require minimum sizes on certain shapes @75%) A key should not be immediately overhead either. It could develop a tendency to slip, place keys @ 15-45° offset from centre.

Figure 07

Figure 08: The structural bond of row 2 is offset (even at the expansion joint) to Mitigate potential bypass.
Figure 09

Sufficient bracing must be done throughout the installation to ensure a tight fit between linings, no sagging is allowed. Should a new lining be tied into an existing already moved lining, it must not be built to compensate for already cycled linings. This is a challenge on its own as quality workmanship is required. More often than not this is managed by backfilling worn intermediate linings and / or kicking the old hot face lining out to its original position having replaced keys. If a step in the lining is inevitable – having it with the flow as oppose to it being an impact area would have to be the goal.

When a partial reline is planned demolition must be done in an offset fashion to avoid straight joints during installation. Meaning that each ring, course and layer of refractories should at least be stepped by 100mm or whatsoever the relevant specification requires.

• • Tube sheet & Ferrules:

I found that some client needs ferules to complete a project and whatever is available was deemed fit, considering that a ferule should be a ferule. I hope to clarify some technical aspects that require consideration. 

Since the tubes are subjected to high temperatures while the boiler drum is subjected to cold water feed there is a natural difference in the amounts of expansion between the drum the tubes and the tube sheet. This being one of the reasons for the tube sheet being refractory lined to ensure a regulated temperature reaches the tube plate. Hot gas exposure also needs to be kept of the tube plate and weld surfaces. Other failure mechanisms on metallic surfaces are discussed in section 8.

The tube sheet is generally a 50-135mm high alumina castable with values equivalent to the shaped alumina’s 90 – 95 noted in Fig 3. Single or dual-layer linings depends on the design where some single layered linings have no anchors, dual layered are either welded or bolt-on stud/anchors. Each design considers restrain limitations of the lining on a tube sheet as there may simply not be space for anchors between ferules but where cycling and any resulting flexing is not ideal it could in cases be compensated for in the peripheral expansion and support.

Tube sheet ferrules may be Inconel, stainless or ceramic which extend between the surface of the refractory lining, over the weld and well past the entrance of the tube. Sometimes they are placed with an intermediate collar resting against the tube face. Should there not be a collar it could be created with a layer of ceramic paper covered with waterproof tape. Unfortunately, it can still slide in the sleeve created. Thus, it is best to ensure the design has the stop end collar fixed. No collar or stopper means it can easily be casted over the ferule face or if the ferule is not sealed into the tube.

Fluted or slightly flared ferrule tips can help prevent a straight entry point for gas along the outside of the ferrule. The collar up towards the tube face also assisting in keeping the process away from welds. Using metallic ferrules in the hot side of a WHB has proven the better option given they withstand thermal shock better and does as a result does not collapse in on the tube. From a feed preheater or cold side perspective the thermal stresses are significantly less and ceramic – as a result cheaper – ferules are valid.

Some ferrule designs incorporate a collar mid-way over the length thus embedded in the cast or dual layer partition area. If necessary, ceramic fibre paste may be used to level out any deviations in the tube sheet plate to ensure a level surface. The ferrule to tube sheet interface must be smooth and a ceramic fibre paper gasket should be placed between the two to ensure a flush fit against the weld seam where applicable. Designs may vary and in some instances the tube is recessed from the plate which will require its own unique approach.

Should the tube face allow for Ceramic fibre in a cavity around the ferule it should be filled. Optionally Ceramic fibre mastic can be used. 

Modular ferrules:

Some tube sheets are not cast; modular ferrule headers are used instead. These shaped headers must line up on all corners and any gaps between (and behind) ferule headers must be sealed with ceramic paper and High Alumina mortar. When a ferrule header does not fit it should be cast/rammed around the ferrule itself.

If in fact the tube sheet is lined with modules it goes without saying that risks related to casting is removed, and time saving can be significant.  Improved localized maintenance and inspection opportunity while avoiding jackhammering adds additional value!

Also considering that modules can shift/move independently on a flexible tube face removing a significant degree of risk related to monolith cracking Movement in the tube sheet that takes place during thermal cycling and flexes the tube sheet often cause of strain on retaining systems – where applicable & cracking of ferrules and monoliths. Once a crack develops, it allows the hot process gases to reach the tube sheet and/or tube. Boiler feedwater (BFW) leakage into the hot refractory lined unit causes additional damage

Typical Ferrule functions are primarily as follows:

1. Protect the tube sheet inlet & tube-to-tube sheet attachment welds from the process carburization/metal dusting and sigma phasing.

2. Affect a smooth flow of turbulent hot gas. 

3. Prevent possible damage to the refractory cross-section caused by erosion.

4. Act as an anchor system for castable (non- modular).

5. And to a lesser extent to mitigate Heat Flux caused by steam potentially trapped at the tube to header interface.

Tube sheets are cast from bottom to top using well-managed shutters, jacked, and supported at the correct dimensions to avoid overcasting (Figure 10 Blocked tube openings). It is worth noting that the formwork should be pre-fit for a dry run ensuring they are dimensionally correct, as previous formers are often used and may either be missing sections or be deformed. In the case of new formwork, it still applies, they should in fact always be pre-fit.

Figure 10

Kicked formwork, laminations, or any weak spots for potential gas by-pass between or within the cast lining should be avoided. Wet on wet or continuous casting to avoid laminating also ensure a homogenous bond which in effect mitigates bypass helps but… 

…that said each OEM has its preferences and for certain tube sheets, controlled construction cold joints may be required to allow a degree of flex. These construction joints could be painted with a water-based paint to ensure no-bond in cases to allow flexing of the tube face and subsequent refractory movement.

5.6 Manway or inspection port:

It has been important to note with some installers that this section is considered less important however if – in the event it’s a pre-cast plug – and it collapses it warrants a shutdown to correct it. If could and is- also gunned or cast on certain cases. I noted that lower grade ceramic fibre products are often used as opposed to Hi alumina grades – in spite of the fact that its specified contactors often mix it up not knowing the implication or reason – thinking it’s the same thing to use Hi Silica fibre. The principle of Leaching Silica still applies even though there are specific areas considering pressure and temperature that may induce it more than at a manway exit in for example the cold side.

6. Expansion:

 Joints for expansion are usually placed at or close to corners or offsets and other changes in the surface plane to absorb thermal expansion and to limit stresses generated in the lining. The necessary provisions should be made depending on the expansion characteristics of the refractory at operating temperatures. In practice only half the theoretical expansion allowance is sufficient, the remaining half will be absorbed by the mortar and dry joints where oil papers have burnt away.

If wool is used expansion joints must be filled with compressed ceramic fiber through the back doglegged or tongued joint. Material expansion characteristics vary as do mechanical attributes of each vessel and related designs may vary but a joint placed 15mm wide every 1.2 meter is a good guide in the temperature environments typically found in WHB’s. I prefer if expansion joints are taped up or closed off with sticky tape to avoid damage or contamination of any kind, such as in & out traffic, water, dust when loading catalyst from above or any possible spills.

 7. Dry out: 

 Before any dry out make sure that all the brick joints are sealed with a high-grade alumina mortar. “Thumb spots” tend to create little cavities from when bricks were placed. The smallest cavity is a potential spot for erosive wear. Naturally, the chamber must be free of any tools, shutter board, and all debris or waste. Wood plugs below for example the diffuser must not be burnt off! Remove all combustibles.

The WHB must be well blocked off below the gas diffuser with Ceramic wool to contain heat during the dry out (if an auxiliary burner is used from the unit manway and not the ATR itself) and the drainage valves must be left open. If drying occurs under rising pressures or operating pressures the heat up curve must consider the variance related to moisture vapor points not being the same as at atmospheric conditions. See figure 11. The worst would be if a unit trip occurs. Water vaporizes from 215°c @ 20 bar vs 100°C atmospheric pressure. A sudden vapour expansion will result in an explosion considering the immense amount of energy in steam.

When a burner other than the ATRs start-up burner is used one must make sure it is installed correctly for the controlled heat up. It is important to verify that the burner tip is installed straight and that the flame does not impinge the lining. 

A pre-conceived dry out schedule must be drawn up and agreed to by the dry out team the client the OEM and the relevant quality assurance party as being sufficient and safe. Allow for small temperature hikes, lining configuration dependent it could be from 15 to 25°C per hour. The equipment could be lagged to contain radiant heat, but temperature couples should be placed or monitored from strategic points. Consideration must be given to the fact that heat rises, and the temperature couples mounted overhead will record higher readings quicker than the lower laying thermocouples. But it all depends on whether the WHB is isolated for localized drying or if there is a mechanical draft from the ATR through to the Feed pre-heater/cold side.

Shifts upward at hold points should only be made after the last or slowest thermocouple has achieved the desired temperature and it has itself soaked at the hold point temperature for the desired time.

One should not assume that the schedule is accurate, thermal conductivity can be higher or lower than expected. Thus, regular checks for vapor escape are necessary. Always make sure the thermocouples are all mounted correctly do not mistake one for another’s readings. 

Drying out multiple linings require a clear understanding that dry pressed bricks do not require dry outs, only the monolithic behind the brick lining. Vapour must not or should not be able to escape through the brick joints; it may escape through expansion allowances and the open drainage system. 

If at any time during the schedule actual temperatures are behind the planned rate of temperature rise – it must not be accelerated, the schedule must be re-plotted to reflect the current situation. It is critical to understand that entrapped steam has a great amount of power and will cause tremendous asset damage if not regulated with a good dry out plan.

Finalizing a dry out requires temperatures to be dropped in a controlled fashion; a safe drop is 50°C per hour. During a post dry out inspection one should look for signs of cracks, dislodgment, discolorations, and explosive spalling. Cracked monolithic linings are a reason for concern, if the dry out went to operating temperature and cracks are above 1mm precautions should be taken during operation. Crack propagation and gas bypass become a likelihood in a lining that cracked before seeing any process conditions, it is a reason for rejection. Post dry out inspection also includes checking whether expansion joints moved correctly over the full circumference, issues in movement here must be addressed and clearly noted.

Figure 11 (see Steam Chart Pressure vs Temperature – at end of paper)

 

In the event a waste heat boiler is flooded after dry-out or having been in service already, one may consider that the non-reversible chemical hydration processes has already occurred. Or for that matter those hydraulic bonds have been achieved through the cross-section and ceramic bonds may as well already been achieved.

It would only be required for a raise to roughly 110°C atmospheric pressure and hold for an hour per inch – lining density and configuration dependent, to remove “free water”. If in the case dense brick is covering bubble alumina and water penetration is significant these rates may naturally be adjusted to allow sufficient drying.

8. Types of failures:

8.1 General:

Failures may emit from poorly installed or managed refractories or such related activities, process parameters not adhered to, or even inaccurate control instrumentation. Failure of a Reforming WHB does not only affect the boiler itself but also the entire reforming and Syngas operation. Improper design or operation of a WHB is simply put, costly. The refractory department must be involved with potential changes in design and operational parameters this should be carefully reviewed and managed to assure their reliability and functionality.

The most common refractory failure related problems in a WHB are: 

• • Tube sheet leaks must be mentioned first, severe tube ruptures call for an immediate shut down. Ruptures forcing pressure into equilibrium may cause damage of an explosive nature removing refractories from a tube face.  

• • Hot spots that have gone out of manageable control with air or steam.

• • Excessive cracking or spalling of a lining (thermal and mechanical)

• • Chemical attack or corrosion from process gases.

• • Excessive shrinkage and development of gaps.

• • Anchor failure and detachment of a lining from the wall.

• • Explosive spalling during dry out.

• • Excessive quenching … wet – instead of dry steam quenching.

• • Maintenance blast cleaning: See image at end of paper …… the wrong grit, grade and operator was used to clean tubes.
• • Hot spots could occur through many a means and any severe lining failure may show up on a thermal scan. Means of “managing” them could be through a reduction in feed rates, or as noted above with steam or air (fans/hoses). In some instances, picking up feed rates for increased expansion to allow gaps to close by themselves could apply. Risky, but if the casing is within safe metallurgical/thermal specs it could be considered. But all of this is not ideal and is usually considered a short-term solution while plans are being constructed to mitigate the situation. The reason for hotspots is almost endless as any cause for refractories to fail and thus not contain the process will cause a rise in temperature to the shell. 

8.2 non-process (activities leading toward failure):

Poor installation methods due to a lack of knowledge or skill, inadequate equipment, poor design, or time constraints, some factors related to poor workmanship may not even be noticed during installation, but it can cause premature failure.

Poorly cut bricks, poor pre-cast shape qualities, inadequate application or bonding and sintering of mortar & brick and incorrect keying. Inconsistent or poor mixing of castable (which may even be the case with inconsistent bag weights) preventing good chemical reaction with allowed water. Over vibration of material which can cause segregation and likely premature setting/hardening due to excessive energy in localized areas of the lining. All the above will complicate installation and negatively impact the general lining stability, performance, and quality.  

High temperatures both atmospheric or water, with water supply coming in past heated or steam lines as example. Or extremely cold conditions that can form free water ice crystals in monolithic materials. This leads to poor hydration due to expansion in the matrix, resulting in a poor hydraulic bond. Ice is expanded H2o and it will leave voids. Not to forget that retarded or flash setting could also occur given warm or cool conditions. Poor water quality with PH values outside the 6 to 8 range and high conductivity or contaminant/suspended solid values in plant supplied water affecting setting time and potential material flow characteristics.

Poor means of warming mixing water like using steam lines with oil and rust potentially inside. Wet curing of material before the initial set has occurred can result in brittle material with no refractory value. Efflorescence or the leaching valuable chemistry ingredients can also occur if curing and firing is not timeous.

Poor shuttering methods on a tube sheet causing overcast or closed tubes disrupting gas flow potentially causing temperature & gas pressure to build up. Drying a monolithic lining out incorrectly and poor temperature control during both commissioning and decommissioning will also result in material failure.

8.3 Process conditions (leading to potential failure):

8.3.1 Fouling and corrosion:

Gas side fouling and corrosion problems occur in all the energy-intensive industries. Fouling in boiler tubes occur thanks to the alkaline salts, carbon, and leached silica to mention the main culprits building up in the tubes restricting gas flow, and not permitting optimum energy transfer causing undesirable temperature gradients. In some instances, it also corrodes the structure it settled on leading to leaks. 

If not managed during maintenance, it has critical implications. 

In short, fouling is the deposition of an insulating layer of material onto a heat-transfer surface which in some cases are accompanied by corrosion or high temperature mechanisms as noted in the ferrule’s function as protection section.

8.3.2 Metal dusting:

API 571: “Metal dusting is form of carburization resulting in accelerated localized pitting which occurs in carburizing gases and/or process streams containing carbon and hydrogen. Pits usually form on the surface and may contain soot or graphite dust.” … “… Metal dusting is preceded by carburization and is characterized by rapid metal wastage… )

It a degree of corrosion that forms on the heat transfer surfaces also referred to as catastrophic carburization. This carburization phenomenon may occur on metal surfaces under specific conditions, especially at temperatures between 450°C and 820°C and when in contact with a gas of high carbon activity. Effectively it results in loss of material, in some cases as metal dust, a mixture of metal, carbides, and/or carbon. In severe cases the material wastage can be very fast, leading to catastrophic equipment failure. 

On alloys protected by oxide layers like stainless steel, metal dusting involves the breakdown of the protective oxide layer, transfer of carbon into the base alloy, formation of internal carbides, and disintegration of the matrix. The metal particles generated by the disintegration of the Iron (Fe) or Fe-Ni-(Iron – Nickel) matrix act as catalysts for carbon formation.

Metal dusting can be prevented, or the risk thereof minimized by applying protective coatings especially Al2O3, so-called allonizing or aluminizing. But despite these possibilities, metal dusting could remain a challenge.

8.3.3 Carbon:

The ATR’s start-up phase creates Soot because of poor stoichiometry. Combined with CO vapor condensation it deposits on surfaces (eg: WHB tubes: fouling), on refractory surfaces (especially near the burner on the neck in the ATR) and withing the cross section of linings. It converts to a solid in the 400°C to 650°C temperature range. Depending on the severity of the sooting phase or process conditions, carbon may be found extensively throughout the WHB hot and cold side chamber surfaces as well, it could be prominent in the cold side given the ideal temperature range for it to condensate at.

Carbon deposits can increase the thermal conductive value of a refractory and penetrating carbon can promote cracks and add mechanical strain on systems. The fouling effect on WHB tubes influences heat transfer values and causes flow restrictions affecting boiler functionality. (Refer to Metal dusting as well)

Refractory anchors and the casing in a WHB generally does not see temps high enough for carbon to diffuse onto a metal. The hot area we see around a WHB at the tube face on a thermal scan is due to conduction via the tube sheet not hot gas bypass.

The use of phosphate or sulphur as bonding agents in the castable or mortar material will result in lower apparent porosity as well as permeability and improved resistance to chemical attack. 

In the ATR burner neck section carbon penetration can create tension-based cracks in the refractory. It occurs behind the hot face in the intermediate layer or in other equipment eg: FCCU – reactor it can occur much closer to the shell reducing insulating capabilities even more. Excessive deposition results in critical lining disintegration often resulting in the collapse of the hot face by pressure from behind.  It is important that as little as possible iron is present in the refractory material.

Carbon monoxide will reduce free iron oxides in a refractory to metallic iron (Fe). The metallic iron present in the refractory material, will catalyse the reaction through which more carbon is deposited in solid form in the castable pores. These carbon filaments grow from the iron oxide and the filaments contain the iron that acts as the catalyst for further ongoing reduction. The carbon deposits therefore spread out over a volume far greater than that of the iron spot. This growth is responsible for the disruption of the castable material.

8.3.4 Silica leaching:

Hydrogen atmospheres above 1150°C and steam above 815°C volatilizes Silica in refractory, as silica monoxide gas (SiO). It mixes with Oxygen in the steam and forms Silica (Sio2) dust which settles on tubes surfaces. If there are high percentages of SiO2 the lining will become unstable. The Silica is soluble in steam which then travels downstream and condensates below 400°C, which happens to be on the heat-absorbing surfaces with temperatures at the boiler tubes. Despite velocity / flow this is where a degree of it will settle thus initially reducing its thermal efficiency and start with acidic pitting or caustic gouging. The point is; choose a refractory with minimal to no Sio2 for this application. Also to be considered is that in a carbon monoxide environment lime (CaO) reacts similarly to Sio2 in terms of leaching and fouling.

8.4 Other:

Thermal shock, thermal fatigue and Cyclic stresses generated by temperature cycling may initiate fatigue cracks. Irregular steam quality, pressure or a damaged quench nozzle can be detrimental. Steam should be super-heated thus dry – not saturated wet as I experienced before- yes It vaporizes almost immediately but introduced at high velocities it still has an impact focused on one area of a lining it can/does cause critical damages.

It’s wise to check the condition of the quench nozzle – flow direction and cavities – all around the lining and from every angle, preferably the open the cross section to allow mechanical teams a comprehensive overview. 

Tube leaks releasing steam into the lining has the potential to introduce unique damage mechanisms as well. With minor tube leaks, a raise in boiler feed water is hardly recognized.

A rise in temperature at the thermocouple ahead of the WHB tube sheet may indicate a blockage or excessive fouling restricting flow, restricted flow ultimately means elevated temperatures and pressures as well as increased velocities – as a result erosion in areas available to flow.

Castable linings compressing behind hot face brick linings resulting from inadequate expansion allowance Causes a castable lining to become brittle and permits gas flow behind the refractory lining to the shell.

Lining discolorations have little bearing on the critical performance condition of refractories in a WHB however the following may be said:

– If it is coated in a white powder – Alumina vaporization took place – it is likely an upstream event or from catalyst dust.

– If it is glazed or has a blue colour to it – it has been exposed to very high temperatures.

– When pink crystals are visible – synthetic ruby formation has taken place which is caused by very high temperature and a mixture of refractory and transition metals. Typically level with the ART target tiles or above the tiles.

It might be added that these anomalies are noticed in more detail at the ATR above and more specifically above the catalyst bed adjacent to the target tiles.

If there are black deposits its simply carbon, but localized black spots are typically signs of carbon bursting on anchor tips close to the surface or iron spotting due to sub grade brick or shapes.

This leads to the hidden damages behind hot face bricks. A hot face lining may look great during the initial inspection, but then you remove bricks for localized repairs or to open up an area at the top of a tubes sheet, or around a nozzle only to find that there was significant bypass, carbon deposits or steam blow outs in a nozzle though the cross section. Effectively compromising a material insulating abilities.

Often it is difficult to monitor areas, the top of a waste heat boiler is seldom continuously monitored, and if one is lucky someone may go around timeously to scan it with a pyrometer or infra-red scanner. It leads to periods of time where the most like spot for failure is out of sight and not on a continuous monitoring system such as a thermal station can offer with live readings. 

9. Maintenance issues

Maintenance strategies depend on whether it is time-, condition based, or as operational requirements dictate. It could happen that the required material to repair is not available even though maintenance is a planned event, one cannot exactly determine scope until an inspection has been carried out. Repairs may also have to be done with an equivalent material which simply means it is not the same as the existing material at all. Raw materials differ and the refining process of it never yields the same results.

A proper inspection of the vertical transfer line or tube in an overhead position is often a problem because of the catalyst load above the mushroom. It represents a hazard simply because it supports an extreme weight and failure of diffuser bricks may result in catastrophe.

A spalled brick lining should not be plastered, if there are no issues on the thermal scan then there should be no need for mitigating actions. Plastering simply pops off at the first thermal cycle. It is usually overhead so, a piece of plaster or mortar falls breaks to pieces and ends up inside the tube sheet through velocity in the unit, thus disrupting the flow. If it really requires repair – then replace the damaged area.

Drilling anchors into the precast shapes or brick to retain material is not recommended. It is practiced in areas of other operations but the chance of creating a hermetic seal is at risk given shrinkage and individual materials thermal kinetics. Plus, the introduced risk of the anchor expanding in its foundation with cracking as a very high-risk factor. Cracks and shrinkage simply translate as gas bypass leaving one with the original or worsened condition.

The T-piece section of the transitions area between the diffuser area and barrel chamber usually has differences in movement it is generally found that the t-piece is partially dislodged, fortunately and usually not to a degree that warrant demolition it can be maintained with ceramic paper and fine grain materials, mortars tend to have less mechanical resistance thus a good castable is preferred. Also make sure adjacent expansion joints are capable of functioning adequately.

When a tube sheet leaks the refractory must usually be removed for mechanical repairs to occur, localized repairs on a cast tube sheet are not recommended but, in my experience, can be carried out without concern.

Tying old and new work into each other or replacing the overhead bricks at the tube sheet normally requires the best skill as old and new does not fit well together. The backing is typically also crushed and saturated with carbon deposits. Building up the rings and layers ensuring a hermetic seal is once again a call for a good strategy and competent staff.

 10. Research and development:

Going Green:

With the current “Green mindset” as well as existing and planned clean fuels projects it is difficult to avoid further development in this field from this point of view. One may say it is critical and imperative to the industries sustainability that heat is not wasted, the transformation of energy and prevention of loss thereof should get more attention than it is, however, it is fair to say that it is getting much more attention than in the past which is a good thing. That said some areas where research and development are making strides are:

– Optimizing existing heat recovery systems to reduce heat losses.

– Growing end-user uses and technologies for low-temperature heat.

– Expanding heat recovery throughout market segments.

– Growing waste heat recovery opportunities.

Material:

From a material perspective Ultra low, low and no Cement castable technologies are developing at an amazing rate. I don’t know if we will ever see plastic gunning in the reforming industry – perhaps yes, where water is a problem, but I believe Sol-Gel materials will be a common place material soon. I believe this development will aid improved reliability at elevated temperatures, faster star-ups (shorter shutdowns), improved refractory lining reliability and cost saving techniques. In essence I am saying that it has the potential to outperform brick lining reliability – potentially replacing “old” ways of installation.

To quote an article on research gate:

The principle behind sol-gel bonding is the formation of a 3-dimensional network (gel) of particles that surrounds the refractory materials and which on subsequent heating develop strength by formation of ceramic bonding through sintering. In the present work, four different precursor sol systems have been synthesized, namely alumina, boehmite, mullite, and spinel by wet chemical synthesis. 

Technology:

An exciting part about R&D in the context of refractories is that Industry 4.0 as it referred to is upon us. We are no longer in the iron or stone-age, the steam and electrical ages has been applied to industry in the greatest sense. The computing age which we are shifting through now is causing Artificial intelligence to take us to almost un-imaginable heights.

We have machines capable of carrying heavy shapes and placing them for us, we have lasers measuring units while online, there are ultrasonics that can detect deviations in linings. X rays can tell us what chemical composition and quantities there off are in materials. We have programs that “thinks for us” 24/7 in terms of when issues arise before it is too late or when repairs should be carried out, or when temperatures are reaching dangerous levels.

I am exited to see what is next… are you?

To conclude; we are living in exiting times with constant progress and adaptability to meet our needs.

Thank you for reading.

Prepared by: 

B. Enslin

References:

A Survey of Gas-Side Fouling in Industrial Heat-Transfer Equipment Final Report W.J. Marner J. W. Suitor. / Johnston Heat recovery / Practical Refractories – Dr J.D Hancock / Theory and operation of secondary reformers – Gerard B Hawkins / Overview of Refractory Materials – A. Bhatia, B.E. / Waste heat recovery – Optimizing your energy system-www.alfalaval.com / Waste-heat-recovery / Silica reduction in secondary reformers www.christycatalytics.com / Transparent Energy Systems Pvt. Ltd.India / Waste heat recovery Technology and opportunities in the US industry Prepared by BSC incorporate March 2008 / Waste heat feasibility studies – Confidential / Api 936 Refractory specification / Cannon Hancock Biggs and Associates 3rd Party quality control systems / API 571 Damage Mechanisms Affecting Fixed Equipment in the Refining Industry

Steam Tables Pressure vs Temperature (valvesonline.com.au)

https://www.sciencedirect.com/science/book/9780127999111