What are the Advantages to Different Filtration Technologies?
Compiled by Mike Henley and James McDonald
(Editor’s note: This column is based on recent discussions in the LinkedIn Ultrapure Water and Industrial Water Treatment groups. This column seeks to accurately reflect comments from each contributor. On occasion, there may be the need to edit contributor comments for clarity or length. An important purpose of each group is to provide a forum for practical examination of issues facing endusers of high-purity and industrial water.)
Clint: With regards to quality/economical value, are there any substantial advantages and disadvantages associated with the different UPW filtration technologies? I've come across many different opinions of both cartridge filtration and ultrafiltration (UF) that make it seem as if different applications may call for one filter over the other.
Mike: “Clint, thank you for your question. In terms of economics, I would like to let other group members offer their insights. You are right about cartridge filters and UF being appropriate for different needs. Generally, UF is associated with tighter spec water since it will remove suspended solids down to 0.01 micron. (There are multiple sources on the filtration spectrum. One is from Puretec at http://puretecwater.com/resources/filtration-spectrum.pdf.)
UF will remove suspended solids, but for dissolved solids, atoms and molecules, one must use nanofiltration (NF), reverse osmosis (RO) or ion exchange (IX). NF will go down to 0.001 micron in its removal capabilities. RO and IX can go even smaller.
In water systems, cartridge filters are considered useful for pretreatment before an RO, or for certain types of suspended solids removal.”
Dr. V.K.: “It is correct to state that cartridge filtration is a must before feeding water to an RO system. UF is 0.01 micron, whereas normally a cartridge filter of 5 micron is normally used before an RO. This is mainly to avoid any contamination caused in storage of UF-treated water, which is being pumped to an RO system through cartridge filter.”
Nikhilesh: “The choice of a filter would largely depend on the enduse of the water. If the target is pathogen removal, the choice is UF. If the intended use is prefiltration before UF, the choice is microfiltration (MF). Similarly for prefiltration for RO, the choice could be MF, or a 5-micron cartridge filter. If the target is organic matter removal, MF or cartridge filters would not do the job; UF can help to an extent, but cannot remove all humic substances.
In the case of UF, there is no need for chemicals (coagulants, flocculants, disinfectants, and pH adjustment) — that gives some operating cost advantages. But, my guess is that it does not give any capital cost advantage.”
Sergey: “The choice UF-MF is always determined by application and UPW quality requirements. The main goal of MF 5-micron cartridges is a police function— to protect RO membranes from fine particles that could damage the barrier layer mechanically.”
Andres: “There are many aspects that impact the right choice of the UPW filtration technology. Flowrates, UPW specs needed by production, availability and reliability of specialized maintenance contractors, design of the UPW production train (how many filtration stations from municipal water to UPW), company policies, etc. I wish there were a couple of rules of thumb to solve this issue, but every case is different, and no clear-cut solution is available to determine what to use beforehand.”
Applications of UPW (TDS < 5 ppm)
Our next discussion was launched by Ali, who is a PhD candidate in chemical engineering.
Ali: There are many applications for UPW water that I know about such as the pharmaceutical industries, microelectronics manufacturing, and high-pressure utility boiler water. Are there other places that it can be used? I would appreciate your ideas.
Mike: “Ali, thanks for your question. In terms of high-purity water, the traditional uses have been in microelectronics (semiconductors, photovoltaic, and flat-panel display), pharmaceuticals and life sciences, and power generation (sub- and super-critical boilers). Some side applications include laboratory and specialty uses.
The term “ultrapure water”, also known as “high-purity water” has different meanings. A semiconductor fab would not likely consider power plant or pharmaceutical plant water as meeting their requirements. Each industry has specific quality requirements. For example, in the semiconductor industry, facilities do not want particles, organics or dissolved ions in the water because each can contribute to chip defects. Conversely, in the pharmaceutical industry, one key concern is the presence of bacteria and viruses in the treated water, as well as endotoxins and pyrogens.”
Sergey: “The chemical industry is an additional consumer of “ultrapure water” (ammonia facilities, fine chemicals manufacturing in the ultrapure substances segment, etc.).”
Falk: “Also, in semiconductor manufacturing and optical production, there is concern with bacteria, viruses, or endotoxins. Bacteria and viruses are interfering particles here. Endotoxins and pyrogens are TOC.”
Industrial Water Treatment Discussion
Compiled by James McDonald
Minimizing carbon dioxide
James: How do you minimize the level of carbon dioxide in boiler condensate?
Ahmed: “Reducing carbon dioxide attack is determined by a number of factors that include: complexity, size and age of the system, the quantity and source of condensate returned, and the raw water characteristics. Various mechanical means of reducing carbon dioxide (CO2) attack are available that include: hot or cold process softening, ion-exchange of various types and degasification of feedwater.
Other considerations to reducing carbon dioxide attack would include: severity and location of the corrosion in the system, pressure and reducing stations, the feasibility of remote feeding, and the nature of the process as well as any governmental restrictions and regulations.
Condensate corrosion, whether from carbon dioxide or oxygen attack, is dealt with through the use of two basic types of chemicals: neutralizing amines and filming amines. Care must be taken with the feeding of either neutralizing or filming amines. Overfeed of either to a system that has experienced corrosion in the past can lead to rapid removal of iron oxide deposits. Sloughed off iron oxide deposits can cause blockage of traps and valves in the condensate system and in the deaerating heater.”
Gnana: “Carbon dioxide liberates as seen in these two reactions:
2NaHCO3 + heat « Na2CO3 + CO2 + H2O Reaction 1
Na2CO3 + H2O + heat « 2NaOH + CO2. Reaction 2
Reaction 1 proceeds to completion, while Reaction 2 is only (approximately) 80% completed. The net results are release of 0.79 parts per million (ppm) of carbon dioxide for each ppm of sodium bicarbonate (as CaCO3) and 0.35 ppm of carbon dioxide for each part per million of sodium carbonate (as CaCO3).
Due to high alkalinity, carbon dioxide liberates in boiler water and to avoid this, make-up water should to be maintained around 8.3 and externally by deaeration.”.
Ahmed: “This link to an article by International Chemtex Corp. is informative: http://www.industrialchemtex.com/docarchive/pub/TT-18%20Condensate%20Systems.pdf.”
Steve: “Carbonate alkalinity in makeup is the usual source. The decomposition releases CO2 into the steam, which then condenses into condensate. So an easy way to eliminate this CO2 is to pretreat makeup to remove alkalinity— my preference is reverse osmosis (RO), plus you maximize cycles and reduce blowdown. Also, if possible, use a suitable neutralizing amine. Some boiler chemical treatments contain sodium carbonate as an alkalinity builder. If you are concerned about CO2 in steam, avoid or minimize use of this type of alkalinity builder. You can maximize cycles and reduce blowdown, among other steps.”
Michael: “Here are a couple of my thoughts related to issues we've experienced with high-pressure power generation utility boilers:
- Don't use organic-based amines and oxygen scavengers for chemical conditioning of the condensate and feedwater systems. The organic molecule decomposes to CO2 in the boiler.
- Use a condensate polisher. This is really the only way to quickly remove CO2 that enters the system.
- Consider a vacuum deaerator or membrane degasifier in the make-up water treatment system to reduce CO2 in the make-up water to the steam cycle.
- Minimize air contact in the demineralized make-up water storage tank.
- Leave deaerator vents open.
- Minimize air ingress into the condenser.”
Surjeet: “As mentioned by Michael, organic amine and oxygen scavengers decompose into CO2 under boiler-operating conditions. However, we need these products, especially for high-pressure boilers in a process industry. Even if you are controlling feedwater alkalinity, there will be CO2 in the system because of decomposition of organic molecules.
I would draw attention of this forum towards steam traps (which are generally ignored in a plant). A perfectly working steam trap in steam network can help in minimizing CO2 by constantly bleeding condensate and air/ CO2. There should be periodic inspection of steam traps to ensure that they working efficiently.”
Dilip: “The CO2 level in condensate may be controlled by maintaining boiler water alkalinity to < 20% of boiler water total dissolved solids (TDS). Morpholine or cyclohexylamine dosed in demin water to raise pH also helps in maintaining alkaline pH of steam. Ammonia liberated in boiler by breakdown of morpholine or cyclohexylamine neutralizes CO2 of steam.”
A.Ric: “I would place a limit on the alkalinity in the feedwater by minimizing bicarbonate (HCO3) alkalinity. If an RO unit is used to produce make-up water, ensure the CO2 is removed before entering the boiler.”
Nikhilesh: “I agree with all the comments. The bottom line is not to allow buildup of carbonic acid, which is the breakdown product of carbonate alkalinity in the boiler, condensing with water to form carbonic acid (H2CO3). The answer is to maintain the pH at 8.3 or higher, or apply a filming amine.”
Dennis: “You reduce CO2 in the condensate by reducing it in the feedwater! You demineralize the feed and pass it over a degasifier.”
Tim: “There are several pretreatment options and limited chemical options for controlling CO2 in the condensate. Essentially, it all boils down to minimizing/controlling carbonate alkalinity in the makeup/feedwater. I have dealt with groundwater sources that had extremely high levels of CO2; levels that were so high that the makeup water literally effervesced.
In this latter case, a vacuum degasifier reduced the load on the demin trains. Now that RO systems have become much more viable, when compared to demin trains, the focus should be placed on the pretreatment system. Dealkylizers are still relevant in lower-pressure systems. Chemical neutralization of makeup/feedwater alkalinity contributes to conductivity, consequently threatening energy costs of the operation. An overall cost analysis should be conducted to determine the best means of addressing the problem.
I may be missing something, but if we're dealing with high-pressure/high-purity boiler systems, I don't understand the chemical reactions that liberate CO2 from an all-volatile treatment (AVT) approach. How do you increase CO2 levels with organic molecules that do not contain O2? Unless consideration is not given to the oxygen scavengers and amines used for O2 scavenging/CO2 neutralization, why would this scenario be a concern? In high-pressure/high-purity systems, the amount of organic O2 scavengers and amines are at very low use concentrations.
I can understand the thermal stability of specific organic molecules, but unless there is a poorly performing deaerator allowing >7 parts per billion (ppb) of dissolved O2, how is there a significant increase in CO2 levels? Typically, the amines are used for buffering the feedwater pH and the condensate. Condensate polishers are, from my personal experience, used for preventing iron from entering the feedwater system.
In the pulp and paper industry, it was not unusual to dump the condensate following a paper break (machine shut down) that resulted in flash corrosion and short-term elevated iron throw from the dryers. Typically, we satellite fed filming amines to the lower-pressure steam feeding the dryer cans.
I guess it boils down to terminologies. I have always considered high-pressure to begin at 600 psig, when sulfur trioxide (SO3) becomes an issue because of its volatilizing.”
Robert: “System changes can often lead to unexpected results. System “improvements” initiated to improve system operations can result in catastrophic system failures.
For example, changing pH control (caustic) chemical injection point upstream of the steam deaerator (DA) resulted in CO2 chemically converting to CO3. The carbonate alkalinity passed through the DA as a particulate and entered the boilers. The CO3 undergoes thermal degradation and releases CO2 into the steam and ultimately the condensate system, forming carbonic acid. The resulting condensate system maintenance to repair/replace condensate system piping due to acid attack was in excess of $100K/year until the issue was corrected.
If your DA is operating correctly, most/all of your CO2 should be removed. Another source of CO2 is in the condensate collection/return system. Many of these systems are not pressurized and use collection tanks that are vented to atmosphere. When these tanks are pumped down, air (with CO2) is drawn into the tank and reabsorbed. The colder the condensate, the more O2 and CO2 reabsorption occurs. Effort should be made to maintain temperatures as high as possible (insulation), which also results in better plant thermal efficiency. If system conditions allow, modification/conversion to a pressurized condensate return system should be considered.
A key indicator for CO2 problems is low condensate return pH. Carbonic acid can drop condensate pH to as low as 4.3. If prevention of CO2 reabsorption is not practical, consider removal by GTM (gas transfer membrane) or a vacuum tower as close to the source as possible. With very hot condensate, practical limitations exist for both of these methods.
Neutralizing Amines injected into the steam system will alleviate the symptoms of CO2 contamination, but will not correct the problem’s root cause. Here are some suggestions:
- Eliminate CO2 (and O2) from the make-up water (RO permeate is preferred to remove TDS). If you have a mixed-bed IX or EDI system (typically high-pressure steam generation), removing CO2 prior to these units can pay for itself (chemical usage, handling, and service fees) in a very short time and reap additional benefits in reliability and quality.
- Ensure the DA (or degasification system) is working properly. These systems should eliminate your CO2 and O2 problems if designed and operated correctly. Only minimal chemical O2 scavenger should be needed.
- Minimize chemical usage as much as possible. Throwing more chemicals at a problem is throwing money away.
Understanding how your system is supposed to work and optimizing each step in the process will help you reach your operational goals (capacity, reliability, and quality) and reduce operating expenses.”
Log in or Subscribe to Access the Full Article
To read or download full-length articles you need a subscription to Ultrapure. Please log in or subscribe below.