Technical Papers / Presentations

The backwashing procedure connects the bottom of the carbon filter bed to the influent water port. The carbon media is then expanded approximately 50% if the backwash flow rate follows conventional practice.

The purpose of backwashing the bed is to remove the silt and other sediment that enters the bed during normal service mode. It is also thought to “freshen” the bed by changing the flow path of the water passing through the bed. However, at the same time influent water which has the same silt and sediment is being introduced into the bottom of the bed and because the bed is expanded, it is introduced to the whole bed. This basically contaminates the whole bed for the entire time that the bed is being backwashed: the longer the backwash; the more the contamination of the bed. After the backwash has finished, the bed then is forward rinsed. This is to “set the bed” for service once again and rinse out the silt and sediment from the bottom of the bed. However, the forward fast rinse compacts the bed trapping the sediment in the bed. Any silt that remains in the bed then can then slough off (typically at an expediential decay rate) into the effluent water of the filter. This has been widely observed when chlorinated organics enter the bottoms of the carbon beds when the municipalities change to 100% free chlorine. The sloughing off of the chlorinated organics can be measured for several hours.

Thus, it is critically important to always use the best influent water possible to back wash the bed. This means that the polisher carbon bed must have water that has passed through the worker for the greatest period of time. If the worker bed is backwashed first, then it will be sloughing off silt and contaminates to the bottom of the polisher filter. Again this has been evident in the past in facilities where the municipality has converted over to 100% free chlorine and chlorinated organics were detected after both the carbon worker and carbon polisher at the same time.

Reduce the frequency of carbon bed backwashes to the minimum possible

Carbon bed filters perform two functions – remove chlorine compounds and remove silt and other sediment from the water. Traditionally carbon bed filters have had no effective sediment removal in the water treatment system to protect them. Some water treatment vendors have used multimedia filters thinking that they would provide the protection necessary. However, as multimedia filters only filter down to 40 µm when fresh (and they are maintained fresh by frequent backwashing) they are completely useless in removing silt. As such, carbon bed filters filter out the sediments from the influent water. By removing the sediment before the carbon bed with adequate pre-filtration, carbon bed contamination is significantly reduced allowing it to perform with greater efficacy.

There are many hemodialysis water treatment systems that do not use backwashable heads on the carbon tanks. Many of these carbon beds (typically portable exchanges) are used for over 12 months without any issues. Facilities are encouraged to exchange the carbon beds either quarterly or semiannually. However, those facilities that do not, do not appear to have any problems with the extended use of the carbon beds. This puts into question the necessity of having to “refresh” the bed by constantly changing the flow path within the bed.

The major reason for not backwashing the carbon beds is to reduce the carbon fines and silt sloughing off the carbon bed. During backwash the carbon media is plummeted around the tank in a mixing action. As GAC carbon is abrasive, this causes it to scour the granules. The resulting carbon fines then are fed to the reverse osmosis (RO) machine. Studies have shown that the major food source for the biological life on RO membranes is carbon fines.

In addition, it is interesting to note that the only water treatment systems that do not experience false positives for chlorine breakthrough when the municipality goes to 100% free chlorine are those water treatment systems that do not have backwashable heads.

A further reason for not backwashing the carbon beds is the saving of water that is wasted for an operation that is not only not needed but is detrimental to the rest of the water treatment system.

Prevent backwashing of carbon bed filters during the period that the municipality is flushing out nitrification bacteria

The USEPA allows municipalities to periodically switch from monochloramines (combined chlorine) to free chlorine. This is to flush out the water distribution system of nitrogen-eating bacterial that thrive on monochloramines. When this happens, there is a bloom of dead bacteria that is chlorinated by the free chlorine. This sediment is flushed into the hemodialysis water treatment systems and shows chlorine breakthrough with DPD-based chlorine sensing instruments. Portable exchange carbon tanks should be installed before the carbon bed filters to remove the organics. However, the most important activity is to disconnect the power to the backwashable heads to ensure that the bed is not backwashed during this time.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2016 February 28

North Texas Chapter of NANT 16th Annual Symposium (2015 November)

Hemodialysis Water Treatment Today’s Technology and Trends (Download in pdf version)

Burke A. West, P.E., MWS, CI, WTS-III
Nelson Environmental Technologies, Inc.

2015 November

Nelson's Contributions to The Hemodialysis Water Treatment Industry (2015 February 11)

Nelson Environmental Technologies’ hemodialysis water treatment systems have been providing water to medical facilities since 1994. Nelson received FDA 510(k) clearance for its Nelson Water Systems for Hemodialysis (k993877) in March 2000. Over the years Nelson has made several contributions to the industry.

These contributions include:

Using different port fittings on PE tanks to ensure proper connection
Shrink wrapping PE tank fittings to ensure the tanks have not been tampered with
Advocating the removal of multimedia filters from the water treatment systems. Replace with multi-cartridge filters where necessary
Identifying the cause of chlorine false positives when municipalities run 100% free chlorine
Developing a protocol for facilities to use when municipalities run 100% free chlorine
Advocating the development of test strips for chlorine testing to avoid false positives
Developing a carbon sample port filter
Advocating less water usage in the water treatment systems
Reducing water heater costs by shutting off hot water supply during pre-treatment back washing
Developing a pre-RO sediment filter / UV sterilization combination to control Bio-loading of RO membranes
Developing an in-line self testing primary detection chlorine monitor / alarm system for water treatment systems

Fool-Proof Portable Exchange (PE) Tank Installation

Portable exchange tanks traditionally use the same fittings on both inlet and outlet ports. Usually they were the “male” side of quick disconnect fittings. However, when the tanks were input into service, occasionally they were put in backwards. This cause an almost immediate failure for mixed-bed deionization tanks and either dumping of media or less than optimal results for all other PE tanks. To ensure that the user connected the tanks correctly, Nelson pioneered using a “male” and “female” fitting. This eliminated user error which also saves the customer money. The industry has largely followed Nelson’s lead.

Shrink-Wrap Sealing of Portable Exchange (PE) Tank Fittings

Portable exchange tanks traditionally had no protection against unauthorized opening of the caps and plugs that protected the media inside the PE tanks. Also, the status of a tank, used or unused, was difficult to tell without verifying if water was added to the bed. To ensure that PE tanks were not tampered with or the media exposed, Nelson pioneered the use of shrink-wrap over the port fittings. Other PE tank suppliers to the industry have follow Nelson’s read.

Removal of Multi-media Filters

Nelson has advocated the removal of multimedia filters from hemodialysis facilities since the 1990s. Multimedia filters can be very effective removers of sediment when applied properly and used correctly. Unfortunately, they are misapplied and misused in almost all hemodialysis water treatment systems. With fresh media or after backwashing multimedia filters can filter sediment down to between 20 µm and 40 µm. Then as water passes through them a cake will eventually start to build up in the media. Once the cake builds up the multimedia filter can filter down to the submicron range. Multimedia filters are normally backwashed only after a 10 psi increase in pressure drop across them occurs. The longer the filter is not back-washed, the better. The worst thing that can happen is to backwash the filter every day or sooner than the cake can build up.

Most municipal water supplies have little to no sediment in the 20 µm to 40 µm range. Backwashing every day or once a week ensures that the filter will never remove any sediment.

The use of multimedia filters in hemodialysis water treatment systems is a crime against nature. Not only does it not work in almost all facilities, they are set to backwash daily. Each backwash uses 20 minutes of water typically flowing at 50 gpm or 1,000 gpd. This is 312,000 gallons or 1,181 m³ of water wasted per year. Not only is this wasteful but expensive. If the cost of water and sewage is $10 per 1,000 gallons, the annual cost of the wasted water is $3,120. A multi-cartridge filter with 1-µm cartridge depth filters would pay for itself in less than one year.

The Adverse Effects of Municipalities Running 100% Free Chlorine on Hemodialysis Facilities

In 2001 Nelson, along with Rick Rodrigues of FMC, identified the sudden rash of chlorine breakthroughs at hemodialysis facilities. Unbeknown to the industry at that time, municipalities were given permission to run 100% free chlorine for typically two weeks to clean out the nitrification bacteria in the distribution lines. These bacteria thrive on monochloramines that the municipalities were adding to the supply water. Unfortunately for the hemodialysis facilities downstream, the free chlorine killed the bacteria and sent a bloom of chlorinated organic material directly into the hemodialysis clinic water treatment systems. When the carbon filters backwashed, this material was introduced into the expanded bed and then trapped there by the forward rinse cycle. Once the facility started up in the morning the chlorinated organic matter sloughed off and gave false positives for the DPD type of chlorine testing that was done at the time. Once the problem was identified, Nelson advocated using a temporary protocol to use during the period of time that the municipality was using free chlorine.

Facility Protocol to Implement when Municipalities Switch from Monochloramines to 100% Free Chlorine

The problem obviously was not chlorine breakthrough but false positives caused by the way that the chlorine was being tested. Because most facilities had not encountered this problem before, the protocol in place was to re-bed the carbon tanks when chlorine was detected. Therefore carbon filters were being re-bed even though the facility knew that it the chlorine test was a false positive. Nelson then advocated using a separate protocol to cover municipalities running 100% free chlorine. This protocol included putting two 3.6-ft3 PE GAC carbon tanks in parallel in front of the system carbon tanks. These carbon tanks were not used to protect against chlorine, but to act as a sediment filter to capture the organics in the water. At the same time the backwashing of the carbon tanks would be suspended until the end of the 100% free chlorine period. This was to protect the bottom of the carbon beds from receiving organic material that would be blown past the two PE tanks at the backwash flow rates of the carbon tanks. The protocol that is on Nelson’s website has generally been adopted by the entire industry.

Advocating the development of test strips to test for chlorine in the water

At the same time as the alternate protocol was developed, Nelson advocated for the development of test strips to be used instead of the traditional DPD test. Test strips measure the chemicals dissolved in the water and do not react to the solids in the water. The DPD test measures both dissolved chlorine and suspended organic chlorine. Test strips were developed and now are universally used to ensure that false positives do not occur because of organic solids in the water.

Carbon Sample Port Filter

Before test strips for testing chlorine were developed and put into general use, the chlorine was tested by the DPD method. Unfortunately, it’s testing of chlorine was “global” (included reading the chlorinated organics that were suspended in the water not dissolved in the water). As such, whenever organics were in the water, false positive readings were recorded. This frequently resulted in re-bedding carbon tanks that did not need to be re-bedded. This resulted in wasted expense as well as disrupting the treatments of patients. Nelson developed a 0.1-µm membrane sample port filter for the carbon test water that removed most of the organics. The resulting chlorine readings were well below the limits which allowed the facility to operate and save the cost of un-necessary re-beds.

Water conservation in Hemodialysis Water Treatment Systems

Nelson has been advocating the reduction of water usage since the beginning of the millennium. Water is not only wasted backwashing multimedia filters but also in backwashing carbon filters. Backwashing a carbon filter is necessary for only one reason – to remove the sediment from the carbon bed. Many facilities that have carbon filters without back-wash heads only change out the carbon when the pressure drop from the sediment build up restricts the flow of the water. Periodic back-washing of carbon tanks is a good idea to remove the sediment. However, each time that the bed is backwashed the carbon media is tumbled against itself and causes carbon fines to break off. These fines are then carried with the flow of water to the RO where they become food for the microorganisms on the membranes. Thus, the frequency of the back-washing of the carbon beds needs to be a prudent decision considering the benefits of removing the sediment against the use and cost of the water and the generation of carbon fines. As most carbon bed filters have 7-day times, Nelson advocates back-washing the carbon tanks one a week. For facilities that are back-washing two 24” carbon tanks daily the savings of water per year is 2 tanks x 20 min/tank x 24 g/min x 6 days/wk x 52 wk/yr or 250,000 gal/yr. The annual savings for the water at $10 per 1,000 gallons is $2,500. But more importantly, 250,000 gallons of water or 945 m³ per year will not be wasted. Over a 10 year life of a facility that is 2.5 million gallons of water saved.

Energy Conservation by shutting off hot water during pretreatment backwashing

Many hemodialysis facilities use blended water. The blending is done automatically to maintain a constant temperature. This arrangement works well for service use during the periods of time that the water treatment system is providing water to the hemodialysis machines. However, when the water treatment system gets maintained over night during the off-hours, this arrangement is very costly. The typical service flow rate is around 10 gpm. The system is designed to provide tempered water around that flow. However, during the backwash of the multimedia filter and carbon filters, the flow is typically between 30 gpm and 50 gpm. Also when the softener is regenerated, the backwash and fast forward rinse are typically in the 20 gpm to 30 gpm range. This means that much hot water is being used for the maintenance of the pretreatment beds. All of this hot water is wasted energy as none of the components need warm water for the maintenance.

Since 2011 Nelson has provided hemodialysis facilities hot water solenoid valve shut-offs to help their customers save costs. This is one of Nelson’s products designed to help its customers be good corporate citizens of the world.

Pre-RO Filter/UV System

Reverse osmosis (RO) membrane fouling has been a major industry problem since the beginning of the industry. Membranes have been traditionally cleaned monthly or quarterly with high pH and low pH cleaners. Even with this maintenance, not all membranes have been trouble-free. Two recent discoveries in the water treatment industry have helped shed light on why fouling has continued to be such a problem.

The first discovery was in understanding the scaling problem on RO membranes. Scaling and bio fouling have traditionally been considered different causes of fouling. However, recently microorganism fouling has been identified as the major cause for almost all fouling. There are two basic types of scaling – wall surface scaling and bulk scaling. Wall surface scaling is typical when temperature is involved in the scaling process such as boiler tube scaling. Bulk scaling occurs when the saturation index of a compound is exceeded and the compound precipitates out of the solution in bulk. This is currently thought to be the issue with RO membranes. As the biomass on the membrane surface grows, it captures the precipitated compounds and scale builds. Without the gel from the bacteria colonies, the precipitated compounds would exit the membranes unimpeded.

The second discovery was that the main source of food for the biomass on the RO membranes is the carbon filters in front of the RO machine. When the carbon filters were removed, the biomass did not increase. Traditionally, the only filtration between the carbon filters and the RO was at the input of the RO. These were 20” cartridge depth filters that basically took out the loops in the water.

Nelson, along with Mark Rolston of FMC, pioneered a pre-RO sediment filter/UV configuration to sterilize the bacteria and remove its major food source. The results have been spectacular. In one facility which had been changing out the RO membranes quarterly, after one year with the pre-RO filter/UV combination, the membranes performed like new and when removed from the housings looked like they did when installed. This system has been widely copied (but not necessarily equaled) by the rest of the hemodialysis water treatment industry. Replacing the 5-µm, 20” depth filters on the RO with 0.35-µm pleated filters that are replaced monthly shows the effectiveness of this configuration.

In-line continuous chlorine monitoring/alarm system for patient loop water

Nelson has been developing an in-line continuous chlorine monitoring system for hemodialysis since 2010. The systems use primary detection methods that measure both combined and free chlorine. The devices are self-testing to ensure functionality as the chlorine concentration readings are typically 0.00 mg/l. Once these devices receive FDA 510 (k) clearance, they will be offered to the hemodialysis market.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2015 February 11

North Texas Chapter of NANT 13th Annual Symposium (2012 November)

Operating a Hemodialysis Water Treatment System (Download in pdf version)

Burke A. West, P.E., MWS, CI, WTS-III
Nelson Environmental Technologies, Inc.

2012 November

North Texas Chapter of NANT 12th Annual Symposium (2011 November)

Water Treatment Systems: What You Need To Know: Documentation, Actions, and Alarms (Download in pdf version)

Burke A. West, P.E., MWS, CI, WTS-III
Nelson Environmental Technologies, Inc.

2011 November

Multimedia and Industrial Multi-cartridge Filters in Hemodialysis Water Treatment Systems (2011 February 24)

Sediment filters are often recommended before the RO machine especially if the SDI15 is greater than 5. Sediment filter type, size, location, and maintenance are critical to effective sediment removal. Multi-cartridge filters pre-RO are preferred to multimedia filters.

Multimedia Filters

Multimedia filters have been around for many years and are preferred over older sand filters for sediment removal because of their efficacy and their smaller space requirements. In industry, when properly operated and maintained, studies have shown that they are effective in removing solid particles down to 1 ìm. To achieve these results the flow of the water through the filter is more or less constant and the filter is controlled by a delta pressure switch that only backwashes the filter when a pre-determined delta pressure (usually 10 psi) is reached across the filter.
Filters with new media or newly backwashed media are capable of capturing particles down to around 20 um to 40 um. However as sediment gradually builds up in the filter their efficiency increases and smaller particle sizes are captured providing the water flow is more or less continuous. This process continues until the capability reaches around 1 um at which time the filter usually needs to be backwashed. Shocks by sudden stops and starts of the flow can dislodge the sediment and cause it to be purged out of the bed.

Multimedia filters have been used for years as the first filter in hemodialysis water treatment systems. However, they are basically ineffective as sediment removal filters as they are not usually correctly sized or operated correctly. In these water treatment systems the water flow is constantly being stopped and started and the beds are backwashed either daily or bi-daily. This means that the only sediment that is being filtered out is greater than 20 ìm to 40 ìm. Hemodialysis water treatment systems don’t normally have removal demands in this range considering that algae spores are typically less than 10 ìm and colloidal particles are in the 1 ìm range. For these reasons multimedia filters have not been normally recommended for hemodialysis water treatment systems in the past. However, today many hemodialysis water treatment systems have constant pressure booster pumps incorporated into them which ramp water flow up and down in place of the older systems that tended to produce water hammer with abrupt starts and stops to the pretreatment water flow. When these booster pumps are used, the multimedia filter can become a viable candidate providing it is sized and used correctly.

Multimedia filters are most effective at removing small particles at lower flow velocities. The optimal size for most hemodialysis facilities is a 24” diameter tank. The flow rate at 30 gpm (backwash of carbon tanks) is 9.6 gpm/ft2. This is close to the recommended flow rate of 8 gpm/ft2 for capturing finer particles.

Industrial Multi-cartridge Filter

Multimedia filters are installed at the beginning of the pretreatment section in the belief that they will remove solids such as algae, micro-organisms, other organics, and other solids. The intention is to protect the carbon filers which will also filter sediment out of the incoming water. The ultimate goal is to protect the RO from excessive organic overload. However, studies that David Paul performed have shown that the carbon tanks are the primary food (carbon fines) for the microbiological growth on RO membranes. If there appears to be a history of heavy loading of the RO membranes, one of the most effective ways to reduce it is by using a multi-cartridge filter with 1 ìm cartridge filters before the RO. It is important to place the filter after the carbon tanks. The size of the filter is critical for operating successfully. A large industrial filter, such as a Shelco that is rated at around 80 to 100 gpm, is needed in order to have the reduced velocities through the cartridge filters. As with multi-media filters, it is best to not change the cartridge filters until there is approximately a 10 psi drop across them. As they load up, they become more efficient. If an ultraviolet disinfection unit is installed before the RO the removal of the particles before it will render it much more effective.
Biological fouling of the RO membrane is the most prevalent type of fouling in hemodialysis ROs. This is particularly a problem as microorganisms can grow across a membrane especially if the water is stagnant in the membrane. The pre-RO industrial cartridge filters are much more effective than multimedia filters for reducing RO membrane microbiological loading.

Cost Analysis

Considering a water cost of $2.00/1,000 gal and a sewer cost of $6.00/1,000 gal, the cost of daily backwashing the multimedia tank is $8.00 per day or $2,920.00 per year. This means that the cost of an industrial multi-cartridge filter would cost about 8 months of water usage for a daily backwashed multimedia filter. A delta pressure backwash controller would pay for itself in about 6 months.

Environmental Cost

Using an industrial multi-cartridge filter will save 365,000 gallons of water a year.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2011 February 24

North Texas Chapter of NANT 11th Annual Symposium (2010 November)

Three Major Issues Today For Biomed Techs Regarding Hemodialysis Water Treatment Systems (Download in pdf version)

Burke A. West, P.E., MWS, CI, WTS-III
Nelson Environmental Technologies, Inc.

2010 November

Multi-cartridge Sediment Filters (2010 December 6)

Multi-cartridge sediment filters, such as Shelco, used before the reverse osmosis (RO) machine are one of the most effective ways to keep biofilm off of the RO membranes. When sized correctly and using 1-μm depth filter cartridges, the filter can remove sub-micro-sized particles. This is particularly significant when carbon filters are used pre-RO. The carbon fines that slough off from these filters are one of the prime sources of food for microorganisms living on the RO membranes.

Two factors are critical for proper filtration to take place. First of all, the cartridges must be depth filters. This means that filtration occurs throughout the cartridge, not just on the surface. The second factor is the velocity of the flow through the filter. In order for the filter to work most efficiently, a cake must build up in the filter. This occurs when the flows through the filter are slow enough so that the particles are captured and not blown through. Ideally, this occurs at a significantly less flow than the rating for the filter.

Big Blue vs. Multi-Cartridge

A 20″ Big Blue cartridge has 4.25″πx20″ or 267 square inch of surface area. At a flow of 12 gpm, the flow through the surface is 6.5 gallons per min. per sq. foot.

A four-cartridge housing for 2.5″ by 30 depth filter cartridges has 4×2.5πx30 or 942 square inch of surface area. At a flow of 12 gpm, the flow through the surface is 1.8 gallons per minute per sq. ft. This means that the flow through the multi-cartridge filter is only 1/3 of the velocity of the flow through the big blue for the same quantity of water. This also means that the capacity of the multi-cartridge filter is 3.5 times that of the Big Blue. Based on current pricing, this means that the filter replacement cost for the Big Blue is 2.9 times that of the multi-cartridge.

The lower velocity of the water through the 4-cartridge filter makes it much more efficient in the sub-micron particle size range than the Big Blue at the typical flows feeding ROs for hemodialysis.

Thus, the multi-cartridge filter provides a significantly better performance in the sub-micron range at a significantly less cost than the Big Blue filter.

Bio-film Loading of RO

If bio-film loading of RO membranes is an issue at a hemodialysis facility, it is highly recommended to use an adequately sized multi-cartridge filter before the RO. It will significantly reduce the amount of food available to feed microorganisms on the surface of the RO membrane. Used in conjunction with UV sterilization, it is a very effective method of controlling bio-film growth on membrane surfaces.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2010 December 6

Organic Scavenger Filter Usage In Hemodialysis Water Treatment Systems For High Levels Of Organics In Feed Water (2010 February 25)

Multimedia filters are often used when the municipal water system has high organic content. These water systems are usually surface water fed and often in warmer climates. Unfortunately not only is the multimedia filter ineffective in these cases but it will also give a false security that the problem is being treated.

Usually the higher the organic content in the water, the more chlorine the city puts into the water. When the water is warm, even more chlorine is added. This results in the organic content becoming chlorinated. Unlike multichloramines and other combined chlorine that are dissolved in the water, these are solids. These solids are largely filtered out by the carbon filter and thus are not a problem unless they overload the carbon filter. However, they become problematic when the carbon filters are back washed. Because the chlorinated organics are solids they enter the 50% expanded carbon bed at the bottom during the backwashing cycle (10 minutes typical). Then after a short settling period (4 minutes typical), the bed is forward rinsed thus compacting the bed. During this cycle some of the chlorinated organics are washed out. However, often not all of the solids are sloughed off during this period. Those that are not will gradually be sloughed off during the service cycle. Before the water treatment system is used each morning it is run for 15 minutes to test for chlorine. It is exactly at this time that the sloughing is occurring at its greatest rate. The DPD reactant used by most facilities reacts to any chlorine including chlorinated organics. Since both tanks are backwashed a false positive is recorded for both tanks.

For these facilities where high organic content is a continuous problem, an organic scavenger filter can be justified. AAMI RD62 (2006) recognizes the value of organic scavenger filters when used before carbon tanks. The organic scavenger is basically a softener except that the media is a special anion to adsorb organics.

The organic scavenger tank needs to be large (24” diameter) in order to allow for a slower flow of water through the media. The filters need to be able to work at 30 gpm (backwash rate of the carbon tanks).

Another option for these facilities is injection of an oxygen scavenger such as sodium bisulfite. In most cases, the cost of sodium bisulfite injection will be less expensive than organic scavenger filters. One of the advantages of the filters is that they do not require additional testing that is necessary when a chemical is injected into the water treatment system. Another advantage is that the filter uses salt which is already used for the softener. There is no other chemical handling.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2010 February 25

Water Distribution Loop in Hemodialysis Facilities (2009 February 10)

For some time there has been some discussion as to whether the water distribution loop is considered by the FDA to be part of the water treatment system for hemodialysis. At issue is whether anyone can install the water loops. If the loops are part of the water treatment system, then they need to be installed by companies supplying 510 k) certified systems in order for the whole system to be covered by that 510 k) certification.

In order to get a clarification the author of the guidelines, Miriam Provost, was contacted. She advised that “although the loops are not specifically mentioned in the guidelines the intent is there.” She pointed out that the guidelines mention auxiliary parts that include the components that are in water delivery systems. Although this resolved the question, the resolution needed to be communicated to the appropriate people responsible for the construction of new facilities with water delivery loops.

AAMI addressed this issue when they published AAMI RD62:2006.

Section 1.2 Inclusions reads:

The scope of this standard includes all devices, piping, and fittings between the point at which potable water is delivered to the water treatment system and the point of use of the treated water. Examples of components included within the scope of this standard are water treatment devices, on-line water quality monitors (such as conductivity monitors), and piping systems for the distribution of treated water. Also included in the scope of this standard is the quality of water used to prepare dialysate, to prepare concentrates from powder at the dialysis facility, and to reprocess dialyzers for multiple use.”

During a recent FDA audit, the inspector advised that the AAMI RD62:2006 standard should be used today in Texas.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2009 February 10

Procedures For False Positive Reading (2008 November 7)

Background

For many years now in the United States, the disinfection formula of municipality water supplies has been switched periodically from a combination of free chlorine and chloramines to 100% free chlorine[1]. The purpose of this is to flush out the bacterial in the municipal water delivery system. This is an occasional two-week procedure permitted by the EPA. It is required as nitrification bacteria (that eat the monochloramines) proliferate in the distribution system of municipal supplies. When shocked with 100% free chlorine, these bacteria are killed and flushed from the pipes.

Usually, the municipal water works department advises hemodialysis facilities connected to their systems of this procedure so that they can prepare for this change. This is important because there is typically a bloom of chlorinated dead bacteria (and their body parts) that enters the hemodialysis water treatment systems. As these are solids typically in the 1 micron size range and smaller, some of them are filtered out as they pass through the carbon filters. The rest continue downstream to the reverse osmosis (RO) machine. Being solids, they pass through the RO machine and out through the concentrate stream[2].

Hemodialysis facilities need to be aware of this change in chemistry from the city as the chlorinated bacteria in the water give positive chlorine readings when the water is tested with non-specific chlorine indicators such as DPD[3] These readings are referred to as “false positives” because the chlorine indicators are reacting with the chlorinated solids[4] in the water verses measuring actual chlorine in the water itself. If these solids are filtered out of the water, the water will test negative (0 mg/L of chlorine) or significantly reduced. However, the filter pad through which the water was filtered will test positive.

Hemodialysis facility protocols require that if a positive chlorine reading is noted after the first (worker) carbon filter that chlorine must be tested after the second (polisher) carbon filter. If positive after the second carbon filter, the facility is either shut down if the reading is over 0.1 mg/L or goes into reactive mode to correct the problem if less than 0.1 mg/L is noted. Because the readings are in reality chlorinated bacteria and not chlorine, positive readings can be noted after the first and second carbon filters at the same time or in quick progression. Knowing that the city is changing the chemical composition of the disinfectant allows the facilities to prepare for and work proactively (instead of reactively) in order to keep the facilities operational and their patients safe.

Recommended Changes in the Water Treatment Protocols for Hemodialysis Facilities during Municipal Distribution System Flushing

Be proactive when advised by the city of the disinfection reformulation. All of the personnel associated with the hemodialysis water treatment system need to be aware of the pending change. When a false positive occurs, it will not be a surprise and protocols to operate under these conditions should be in place.
Test RO permeate water for chlorine. A false positive can mask a true chlorine break-though. If a positive chlorine reading is made, one concern is always to verify that it is a false positive and not true chlorine break-through. If the facility does not have an in-line continuous chlorine monitor, the easiest way to verify this is to test the RO permeate water on a regular basis during the municipal flush period.
Do not backwash carbon filters for the duration of the time that the city will be on free chlorine (usually two weeks). The reason not to backwash the carbon bed filters is that during the backwash cycle, the dead organic material is being flushed up into the bed from the bottom of the tank. As the bed is expanded, the whole bed is contaminated. When the bed is then forward rinsed, the bed is compacted and the contaminants are only slowly flushed out. Thus, when the system is tested in the morning with a non-specific chlorine test such as DPD, a false positive reading occurs. With further use, this reading will continue to drop. Unfortunately, when the apparent breakthrough occurs, some facilities will quickly re-bed the tanks only to have the same problem in the next day or so. This is why facilities will note chlorine breakthrough in both the worker and polisher filters at the same time but see no chlorine in the RO permeate water.
Add portable exchange (PE) carbon tanks before the worker carbon filter. The 12 by 40 mesh, acid-washed granulated activated carbon makes an excellent mechanical filter for the organics in the water during the municipal delivery system flushing. Thus, much of the organics are going to be filtered out of the water in the carbon filters. Placing PE carbon tanks in front of the worker carbon filter will protect it from receiving the brunt of the contamination. It is important to install these PE tanks before the city changes the chlorine formula so that the worker carbon filter is protected from the start.

Summary

Hemodialysis personnel responsible for operating or servicing the water treatment system must be prepared for those times when municipal water treatment supplies are shocked with 100% free chlorine. The facility must work proactively from the time at the city advises that the disinfection formula will be changed until the process is over. Special temporary protocols should be posted so that everyone connected with the water treatment system is aware of changes. PE carbon tanks should be put in place to protect the worker carbon filter and the back-washable carbon filters should not be back-washed during the period that the temporary protocols are in place. If chlorine break-through is detected, verify that the reading is a false positive by verifying the water flowing out of the RO (both permeate and drain). If the permeate water is chlorine free, the reading is a false positive.

[1] See technical article “Apparent Chlorine break-through Problem Recently Experienced in Texas in the Technical Papers section on website NWS.bz

[2] If chlorine break-through is observed in rapid succession after the worker and polisher carbon tanks, the best way to verify if the positive readings are from chlorinated solids in the water is to test the product and drain water from the RO machine. If the product water tests negative and the drain water tests positive, the positive chlorine readings are from solids in the water. If both streams test positive, the positive chlorine readings are most likely from chlorine (or other oxidant) dissolved in the water.

[3] DPD is non-chlorine specific. It will indicate on any oxidant in the water such as free chlorine, chloramines, ozone, etc.

[4] Chlorinated solids should not be confused with chloramines which are not solids but compounds dissolved in the water.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2008 November 7

Chlorine Concerns In Hemodialysis (2008 July 12)

Background

Chlorine is an essential part of the municipal water treatment process to control microbiological growth. However, it is a danger to the health and safety of hemodialysis patients and, as such, is always a major concern to all hemodialysis facilities.

Chlorine introduced into the hemodialysis water treatment system through the municipal supply water must be removed completely to ensure the safety of the hemodialysis patient. To ensure this the FDA and all States have regulations in place and hemodialysis service providers have systems and protocols in place to ensure the welfare of the patients. However, the operations of hemodialysis facilities are run by nurses and technicians and these individuals have varying degrees of training and competence. Great effort is taken to ensure that mistakes are not made through human error. Murphy’s Law states that if it can happen it will happen. In hemodialysis, even with all of the safeguards in place, patients occasionally are subjected to chlorine in their treatments.

Chlorine is becoming more of a problem due to recent developments in municipal supply systems. Initially municipals disinfected their water with free chlorine. After it was recognized that free chlorine in the water reacted with organics to produce trihalomethanes (THMs), a carcinogen, they started to use monochloramines in order to reduce the amount of free chlorine in the water. This created a major problem for de-chlorinating the water because carbon which has traditionally been used for that purpose reacts almost instantly with the free chlorine but monochloramines do not react with the carbon but are adsorbed by it. The adsorption process is time dependent and requires much more contact time to ensure that it is totally removed. To make matters worse, the monochloramines feed nitrification bacteria in the municipal delivery system and as such, municipalities will on a periodic basis change to 100% free chlorine to kill the nitrification bacterial in the lines. Each change of chemical composition in the municipal supply system is problematic for hemodialysis water treatment systems. Also, there appears to be a trend of increasing the level of total chlorine in the municipal water supplies particularly in the warmer parts of the USA as well as when local flooding is deemed to possibly be a threat to the feed water.

Sodium Bisulfite Injection

For all of these reasons, many hemodialysis facilities have taken the proactive step of protecting their patients with an additional layer of protection. They use sodium bisulfite to remove the total chlorine from the influent municipal water at the beginning of the hemodialysis water treatment system.

Systems such as the Nelson Water Systems for Hemodialysis sodium bisulfite injection system are used to react with the chlorine, chloramines, and other oxidants in the incoming supply water whenever the incoming chlorine level exceeds a threshold level. Sodium bisulfite (Na SO³) is an oxygen scavenger (reducing agent). On contact with other compounds it will rob any available oxygen to become sodium sulfate (NaSO⁴) which is stable in water. Thus, sodium bisulfite has an extremely short life in the water treatment system.

Sodium bisulfite treatment of supply water has been recognized and used for many years in the pretreatment for Hemodialysis water. It is included in Nelson Environmental Technologies, Inc.’s 510(k)-certified water treatment systems (k993877) and reducing agents are discussed in the NANT Water Treatment for Hemodialysis, 1989 Edition, page 124.

Although no one can dispute the efficacy of sodium bisulfite for removing all oxidants, whenever something is added to a process, its benefits and risks need to be evaluated to ensure that there are no significant risks for the added benefits. With over 10 years of experience in supplying sodium bisulfite injection systems to hemodialysis facilities, Nelson Water Systems for Hemodialysis has recognized the risks and developed safe systems not only to the patient but also the facility staff.

Patient Safety

The sodium bisulfite system is placed in front of the carbon filters. This is done to remove the total chlorine before the carbon tanks thus protecting them from excessive levels that could prematurely exhaust their capability of removing these contaminants. As mentioned above, sodium bisulfite is extremely reactive with any oxidants in the water and is quickly neutralized into sodium sulfate. Sulfates are one of the most common compounds in water and have an EPA secondary MCL of 250 mg/L for municipal water. Excessive levels of sulfates can have a laxative effect on the human body.

Sodium bisulfite injection levels are not critical. For economic reasons, it is recommended to inject only that amount that is needed to remove the total chlorine. Theoretically this is typically in the area of 15 mg/L to 20 mg/L. However, many facilities inject up to double that amount because of other oxidants that interfere with the reaction. Excessive amounts of sodium bisulfite have in the past been injected (between 10 and 20 times typical amounts) into hemodialysis water treatment systems without any affect to the quality of the reverse osmosis machine (RO) permeate water.

Whenever anything is introduced into a hemodialysis water treatment system, it is prudent to verify that it has been removed from the RO permeate water. As such, whenever sodium bisulfite is injected into the pretreatment section of the hemodialysis water treatment system, the RO permeate water needs to be tested for sulfite ions on a periodic basis. Hach sells a field test kit suitable for this protocol. On a continuous basis, RO permeate conductivity monitoring will alert if any problem in the RO should for any reason allow excessive ions to pass through it.

Thus, there is no significant risk to the patient for the benefit of significantly reducing the chance of chlorine poisoning. If an additional degree of certainty is desired, an in-line chlorine monitor can be added to the hemodialysis water treatment system before the RO unit. The in-line monitor would then detect any level of chlorine before the RO and alert the facility through a remote alarm at the nurses’ station. (The RO unit will only remove 40% to 60% of the free chlorine. Mixed-bed DI media will remove 100%.)

Facility Personnel Safety

Sodium bisulfite is a chemical and as such must be handled correctly and treated with respect. Sodium bisulfite is an oxygen scavenger, which means it reacts with oxygen and removes that oxygen from its environment. The major safety issue when handling sodium bisulfite is the possibility of inhaling the fumes. The sodium bisulfite removes the oxygen in the lungs and causes asthmatic symptoms – one gasps for more air as the oxygen has been removed from the air in the lungs. Although this is a temporary phenomenon, it can be scary to the one experiencing it (especially when unexpected). Fumes will occur whenever sodium bisulfite comes into contact with the atmosphere. As such, the Nelson Water Systems for Hemodialysis sodium bisulfite systems do not require facility personnel to come into direct contact with the chemical. The chemical is received at the facilities in 5-gallon jugs. The feed to the injection system is through a flexible tube that passes through the cap of the jug. When the jug in service is empty, the cap is removed from that jug and placed on the new jug. The cap of the new jug is placed on the jug that was just taken out of service. The only exposure to the atmosphere is the time it takes to remove the cap from the one jug to the other. In addition to not having to handle chemicals, by using 5-gallon jugs, the facilities do not need to have all of the EPA issues of storing on-site hazardous materials which would be necessary if they used 55-gallon drums.

Summary

In summary, the Nelson Water Systems for Hemodialysis sodium bisulfite injection systems have been developed over many years to ensure a minimum risk to both facility staff and patients and provide a significant benefit of reducing the risk that patients are not exposed to chlorine in their treatments. Additional safety, if deemed necessary, can be provided by in-line monitoring of chlorine with a remote alarm at the nurses’ station.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2008 July 12

Fundamentals Of Water Treatment For Hemodialysis (2002 November 2)

Water treatment systems for hemodialysis have been around for many years. However, the pace of change has significantly increased over the last few years. Higher industry standards, changes in State laws, and better-informed customers have created an exciting dynamic market for water treatment companies today.

Typical Water Treatment System
The basic components of a typical water treatment system for hemodialysis are shown in Figure 1. The blending valve through the softener is called the Pre-treatment section. Equipment from the repressurizing pump to the end of the water treatment system is called the post treatment section. A typical system includes pretreatment, reverse osmosis, storage tank, and post treatment sections. The sections are integrated, often connected together by a master control box. An alarm system ensures that technicians and nurses in the patient area are aware if the system needs attention.

Pretreatment
The pretreatment section serves three purposes:

Isolates the system water from the rest of the influent water system;
Prepares the water for the reverse osmosis (RO) unit; and
Removes oxidants that are detrimental to patient health.

The typical system includes:

A back-flow preventer;
Twin carbon tanks;
And the additional equipment required for the RO unit:

Temperature between 2oC (35oF) and 30oC (86oF);
Pressure between 1bar (15 psi) and 5 bar (73 psi);
pH between 3 and 11 (recommended 5 and 8);
SDI₁₅ less than 5;
Hardness less than 50mg/L (3 gpg);
Chlorine-free.

The first equipment in the pretreatment system is often the temperature-blending valve. It is an automatic thermostatic valve that blends hot and cold feed water to maintain a selected temperature. The circuit includes by-pass valves to be used when blending is not desired.

Back-flow preventers ensure no cross links between the water system and the municipal system occur if the supply system should draw a vacuum. They must meet local plumbing code and be inspected by an authorized inspector before being put into service.

A booster pump is generally needed as few water supplies have sufficient pressure to ensure an adequate pressure reaches the RO unit after passing through all of the pretreatment equipment. The flow and pressure requirements of the system must be well understood and then a pump must be selected with the performance curves that best meet those requirements. The system must include all of the associated valves and hardware needed to use it efficiently. Preferably, a flow switch controls the pump. This arrangement insures adequate pressure not only when the RO is working but also when the pretreatment filters are backwashing. A hose bibb is commonly included after the pump.

Sediment filters typically follow next when influent water dictates their use. The quality of the influent water (particle size and distribution) will determine the best system. Back-washable multimedia bed filters, cartridge filters, and microfilters may be used. If surface water colloids are a problem, a coagulant may need to be injected before the multimedia filter.

The primary reason for carbon filters is to remove chlorine, monochloramines, and other oxidants from the influent water. The Association for the Advancement of Medical Instrumentation (AAMI) Standards have been adopted by the industry. Based on these, most States now require a minimum empty bed contact time (EBCT) of 10 minutes. At least two filters must be used and the last one must have a minimum EBCT of 5 minutes. Acid-washed, 12 by 40 granulated activated carbon (GAC) with a minimum iodine number of 900 is used. The GAC filters also remove sediment as it passes through the media. This is beneficial for the RO unit. However, over time the sediment will clog the micropores of the media and require its replacement. Although frequently-exchanged portable GAC tanks can be used, the most common practice is to use bed filters with control heads to backwash the media either daily or bi-daily. This is usually an economic decision.

The carbon section includes sample ports in order for the staff to test for chlorine at the beginning of every shift and pressure gauges to monitor the filter beds. If high amounts of chlorine or other oxidants are present in the water, an oxygen scavenger such as sodium bisulfite may be required. This is injected before the carbon tanks and will react with the oxidants on contact. This removes them and allows the carbon tanks to polish the water.

One problem that is plaguing water treatment operators today is caused by municipalities changing the disinfection formula in their systems without previously advising the hemodialysis facilities or the water treatment companies. When municipal water supply companies super-chlorinate the water or change to a high dosage of free chlorine in order to control a bacterial problem, the result can be almost immediate chlorine breakthrough of the carbon tanks. In addition to this, the resulting chlorinated organic material, small amounts of naturally occurring multi-chloramines, as well as other halogen compounds that are oxidizers will give false positives after all of the carbon filters. Before this phenomenon was recognized, companies were spending considerable amounts of money rebedding carbon tanks and adding additional carbon tanks. Clearly a chlorine-specific (as well as monochloramine-specific) test kit is needed. Until then, the best approach appears to be to add a chemical injection system using an oxygen scavenger such as sodium bisulfite to protect the carbon tanks (as well as patient safety) during these periods.

Hard water is a problem in many locations. Water softeners are used to remove the calcium and magnesium ions that account for most of the hardness. They will also remove small amounts of iron. In those areas where iron is a problem, iron filters may be needed to protect the RO membranes. Other locations with high alkalinity water may require a dealkalizer. The productivity (permeate flow) of the RO unit is adversely affected by higher alkalinity water.

Reverse Osmosis and Storage Tank

Current AAMI standards require that 90% of the mineral salts called total dissolved solids (TDS) be removed from the influent water. This can be done by reverse osmosis or by demineralizing with deionization (DI) media. For economic reasons, RO units are almost universal. Some customers specify both RO units and polishing with DI media to remove virtually all of the TDS.

The incoming water quality will determine the recovery rate (the amount of permeate water compared to the amount of feed water needed) of the RO. RO units are set usually to recover between 50% and 80% of the feed water. The recovery rate varies proportionally to the cleanliness and inversely to the TDS of the incoming water: the cleaner the water, the lower the TDS, the higher the recovery.

RO systems can be connected directly to the post treatment section. These are called direct feed systems. The advantage is that no storage tank is necessary. The water returns from the patient loop directly to the feed of the RO. However, this system is not popular as any problem with the water treatment system can interrupt the water supply to the hemodialysis machines. In addition, storage tanks allow a constant pressure and flow velocity to the hemodialysis machines.

Storage tank systems include:
Feed water hose and valves;
Conical-bottomed tank with internal sprayer;
Air breather with sub-micron filter, delivery valves and hose;
Over-flow, level-control system, and loop return circuit (including back pressure regulator, pressure gauge; temperature gauge, sample port and flow gauge).

On an emergency basis, and in some applications where portable exchange (PE) DI tanks are needed or economical, deionization is used. When used, the softener and RO are by-passed and the influent water is obtained after the carbon filters. The de-chlorinated water is then typically fed directly to the PE DI tanks in the post treatment section.

Post Treatment
The post treatment stage polishes and monitors the water going to the hemodialysis machines. It typically includes:

The repressurizing pump;
A flow meter;
DI tanks system;
Conductivity monitor and high-conductivity divert; and
Sub-micron filtration (ultrafiltration).

Customers usually specify the stainless steel repressurizing pump configuration. Some customers want twin pumps that they will control manually. Others want automatically alternating pumps. Other customers opt for only one pump in service with a backup pump that can replace the one in service. The pumps normally have protection against running dry. A flow switch is typically used..

The DI section includes:
Hoses for the PE tanks,
A resistivity monitor/controller (typically 1 megohm) after the first PE tank called the worker or scrubber,
A sample port before and resistivity monitor/controller (typically 20 megohm) and sample port after the second PE tank called the polisher,
A cartridge resin trap,
A sub-micron cartridge filter, and
By-pass valve.

Sub-micron filters, or ultrafilters, are commonly used in post treatment to capture bacteria, bacterial body parts (endotoxins), and any other pyrogens that can affect the health of the patient. Filtration can be by 2.5” cartridges that need to be replaced annually or sooner if the pressure drop across them exceeds 10 psi or by 4” x 40” ultrafilter membranes. The advantage of the larger membranes is that they have two flows (the permeate flow and a trickle flow to drain to wash away the solids that do not pass the membrane). These filters are operated typically at 95% recovery.

The last components in the post treatment section are the conductivity monitor/controller and high conductivity divert. The monitor typically reads 0 – 100 us (although 0 – 100 PPM are still very popular). Most States limit the water to 50 PPM (typically 75 us). The monitor/controller is setup to alarm at a value specified by the customer.

When a high conductivity divert is used, the patient-loop water will automatically divert to drain when the setpoint of the monitor is exceeded. It will return the water to the patient loop once the quality of the water is within specifications again.

Alarm System

A number of alarms alert the operators of problems or potential problems. These can include water temperature out of range, low water pressure, RO unit shutdown alarms, storage tank water level low, high conductivity, low resistivity, as well as others. Part of the alarm system needs to include a remote monitor/alarm unit in the patient area that will alert the nurses and identify the problem. This should include both audio and visual indicators as well as an audio mute button that will silence the alarm for up to 3 minutes. The audio if muted must re-alarm to meet AAMI requirements.

Figure 2. shows a typical basic system. As indicated above, the actual system will be dictated by the influent water and customer demands.

Bacterial Control

No water treatment system for hemodialysis can be designed without considering the control of microorganisms. “Burke’s Laws” state:

In water, bacteria colony growth is inversely proportional to the velocity of the water at that location.
Once colonies establish themselves, they will quickly make their presence known.
The longer their presence is known, the more difficult it is to remove them.

Bacteria secrete a gelatinous substance that protects the innermost members of the colony even against repeated aggressive attacks. Ultraviolet disinfection units can be used before the RO, before the conductivity monitor, and before the storage tank to help control bacteria growth. However, the greatest defense is a carefully designed system. This whole area could be considered for another Water Technologies article.

Conclusion

In many ways, designing a water treatment system for hemodialysis is similar to designing one for any other application. One must:

Define the customer’s requirements for the various water quantities and qualities;
Obtain a water analysis of the supply water to be used;
Obtain the site information as to location, size, available services, drainage, etc. as well as local codes and State regulations;
Review process options and select the best one based upon the customer’s desires (value, level of sophistication, quality, durability, etc.);
Then starting at the end of the process, work backwards always making sure that the up-flow equipment and systems meet the requirements of the down-flow systems;
Verify that the site resources are sufficient to meet the demands of the water treatment system;
Verify that the engineered system meets all of the customer’s requirements, if not;
Prepare alternative designs, then explain to the customer his options and the advantages and disadvantages of each one.

However, there is one significant difference. Water treatment systems for hemodialysis are defined as medical devices. The FDA regulates companies designing and supplying these systems. These companies must be registered with the FDA and must submit their projects to the FDA for certification. The FDA will assign them a 510(k) number. Only after the FDA grants certification will they be able to supply their systems to customers.

The process involved in obtaining the 510(k) certification will discourage all but those most dedicated to serving the market. It is also only the beginning. The company must certify to each customer that the system supplied conforms to the FDA-certified system. Companies need to consider seriously whether they have the conditions and commitment to serve this market. The FDA requires that companies have corporate infrastructure in place that will ensure complete control over all activities of the company, that all aspects of each customer’s application is documented and filed, complaint files are maintained, and that the design and supporting literature can not be significantly changed without resubmitting the project to the FDA. These requirements generate significant documentation and “paper trails” that are not required for other water treatment systems. But, with these safeguards in place, the customer can be assured of safe systems for their patients.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2002 November 2

Method To Determine If Chlorine Breakthrough Readings Are False Positive (2002 May 25)

The following is a simple method for determining whether chlorine break-through readings are really chlorine or are false-positive readings (as have been described in previous reports). If a positive chlorine reading is detected, take a second reading using 0.1-µm filter paper to filter the water sample. If the filtered water is chlorine free, then the first reading was a false positive caused by chlorinated organic matter in the water.

After removing the organics, there still may be some naturally-forming chloramines. These chloramines appear to be a form that is easily removed by the RO membranes. If present, it will be necessary to test the permeate and concentrate water of the RO unit as previously suggested.

If there is chlorine in the permeate water, then the chlorine reading is true chlorine as it will pass the membrane.

The simplest way to perform the test is to use an SDI kit with 0.1 µm membrane pads if available. If not, then 0.1µm filter paper and an inline filter holder (47mm) can be used. Connect the filter paper holder by means of a 1⁄4” flexible hose to the sample port and slowly open the sample port until the required amount of water passes the filter. Make sure that the flow is sufficiently slow as to not blow apart the bacteria on the filter surface. Then test the filtered water by conventional tests.

As several of our customers have had difficulty obtaining the inline filters, Nelson Water Systems now provides Chlorine Port Inline Filter Kits that include the inline filter housing, a 10-pack of 0.1µm filter pads, tweezers, tubing and instructions conveniently packaged in a compact plastic box.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2002 May 25

Need For a Total-Chlorine-Exclusive-Indicator-Field-Test Kit For The Hemodialysis Market (2002 March 29)

Recently the need for a true total-chlorine-exclusive indicator that can be used in the field by hemodialysis facilities has bee identified. This is based on the occurrence of false-positive chlorine readings identified at several hemodialysis facilities in the past and the problem is being recognized with greater frequency.

The problem occurs most often when municipalities change the disinfection chemistry of their water supply. This upsets the organic growth in the distribution systems and can release large quantities of organics in the water.

False-positive readings can also occur in municipal systems where the feed water has high organic compound levels. This most often occurs when the supply water is surface water and/or when flooding has recently occurred.

In both of these waters the free chlorine is combining with the organics to produce chlorinated organics. These chlorinated organics test positive when tested by the standard DPD chlorine test procedure, since DPD test will indicate any oxidizer in the water.

The problem today is that almost all hemodialysis companies require facilities to change out their carbon bed if they get any breakthrough through the first carbon tank. When breakthrough occurs after the second tank facilities are prompted to shut down. This is to ensure the safety of their patients, as chlorine is toxic and can result in hemodialysis.

Carbon filters will remove some of the organics, especially with fresh media as the small pores mechanically filter the organics. However, this filtration is short-lived. Large-bodied organics do not adhere to carbon and the water flow pushes them through.

Unfortunately changing the carbon media is an expensive operation and with the prospect of a very short effective life and can be very uneconomical. There are, however, two alternatives. One is to inject an oxygen-scavenger such as sodium metabisulfite. This product works very well, is inexpensive, and is converted into sulfates almost immediately upon introduction to the water. The chemical is introduction, as close to the beginning of the water treatment system as possible and usually only a few PPM are needed. If there is a concern about residual sodium metabisulfite after its introduction point, it can be easily tested down stream with an inexpensive test kit.

The other option is to run the reverse osmosis unit under these conditions but sample the permeate water every 30 minutes to one hour to ensure there really is no chlorine being masked and passing through the membranes.

In conclusion, neither of these, though viable options, solve the real problem. The real problem is the deficiency of the current DPD test kit to test exclusively for chlorine in the water. In order to resolve this shortcoming a test kit would have to be developed that indicate and measure only chlorine or total chlorine in the water. The Hach Company was contacted to see if they could supply such a test kit. They advised that they were aware of the problem but at that time they had no plans to develop one. Obviously one company in South Texas is not going to not have much leverage with the test kit companies. The whole hemodialysis industry needs to come together and work as a block to motivate the test kit suppliers. If they recognize the purchasing power of the industry, I am sure they will be interested in developing such a test kit.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2002 March 29

Apparent Chlorine Breakthrough Problems Recently Experienced In Texas (2001 October 16)

Periodically the USEPA permits municipalities to change from a free chlorine/chloramines mixture of disinfectants to straight free chlorine for a short period of time (typically two weeks). The purpose of this action is to kill nitrification bacteria in the system that thrive on ammonia that is added to the water through the addition of chloramines. According to Dr. Ying Wei, Assistant Laboratory Chief for the Public Works and Engineering Department of the City of Houston, “Even though the chloramine is a disinfectant, there is always a little amount of nitrification bacteria that exist. When the water is to the customer, due to different , or some pipe problems, or the low use of the water (that means the water stays longer in the pipe), chloramine’s concentration to decrease. So the consequence would be the nitrification bacteria will grow faster. The nitrification process is the bacteria take the ammonia as their source, then forming nitrite, then to nitrate if there is enough oxygen in the water.”

This year many Texas municipalities made the switch to 100% free chlorine to rid their systems of the build up of nitrification bacteria. For the most part, the municipalities advised the hemodialysis facilities well in advance of the change. Those affected were appreciative of the advance warning but were unconcerned. They had been taught that free chlorine was not a problem to remove with carbon filters as the free chlorine reacted with the granulated activated charcoal on contact. They had been taught that it was the chloramines that (acting more like organics) were the most difficult to remove by the carbon media. For these compounds, the carbon adsorbed them (captured them) and sufficient contact time with the carbon was critical.

When the municipalities converted to straight free chlorine, many hemodialysis facilities immediately experienced breakthrough in their carbon filters. It was commonly discovered that the incoming free chlorine readings could not be measured, as they were greater then their test kits could measure. Thus, it was assumed that the carbon filters had been overwhelmed by the free chlorine super concentrations coming into the facilities. The carbon filters were rebedded and put back into service.

Unfortunately, almost immediately, the rebedded carbon tanks were showing signs of break-through. This is when the panic started to occur. What was even worse is that the breakthrough was occurring through not only the first carbon filter (the scrubber) but also the second carbon filter (the polisher) as if they were not even there. The facilities had no choice but to rebed the carbon filters again. Catalytic carbon was used in some facilities, regular carbon was used in others, and redox media was added in others. As far as we know, none of the facilities had to rebed a third time regardless of what action they took the second time. What also was interesting was that not all facilities were hit with the chlorine breakthrough problem.

At this point we knew that something was happening that we did not understand. Rick Rodriguez from FMC and I visited the City of Houston, Public Works and Engineering Department. We met with Dr. Ying Wei. Dr. Wei suggested that the problem might be the bacteria in the lines. Water samples were taken at several of the facilities experiencing problems. The samples were analyzed and showed extremely high counts of bacteria.

Three carbon companies were contacted to ask for their help. None of them could offer any suggestions as to what was happening. I finally contacted Norm Wagner of Calgon who had helped explain other unusual phenomenon in carbon in the past. He suggested that Dr. Wei’s theory was right and that the bacteria were giving false positive tests. He suggested that the next time we have one of these occurrences, we filter the water through a 0.45-µm filter paper and test the water that passed through the filter. If the water shows no detect, then the problem is the bacteria.

I contacted technical assistance at Hach, Inc. the manufacturer of the test kits that we were using for chlorine testing. I was advised that any oxidant in the water would give a positive reading. Thus, the test does not exclusively test for chlorine. The technician, Bob Stewart, indicated that there was no field test kit that would do so. In the laboratory, with specially trained personnel, they could verify the presence of chlorine.

At this point we had a theory that could explain what happened but we needed to make sure it was correct. We were able to do that when breakthrough occurred at another facility. Ironically, the carbon filters had just recently been rebedded. Breakthrough occurred for both filters. Chlorine readings were taken from both the RO concentrate and permeate streams. There was no trace of chlorine in the permeate water but there were elevated concentration values of chlorine in the concentrate stream. Osmonics advised that under the conditions that the RO was operating at the time that the expected free chlorine pass through for the membrane would be between 40% and 60%. The fact that none of it passed the membrane was proof that it was not free chlorine. Instead of changing out the carbon filters again, the RO permeate water was monitored on an hourly basis to ensure that no chlorine was detected. After three days no more chlorine breakthrough was detected and the carbon tanks have been working fine ever since.

What have we learned from this experience? Several things.

First of all, when the municipality advises that they are going to be switching over to full free chlorine (or they are going to use significantly elevated levels of any disinfectant), we need to prepare for the possibility that we will be getting false positive chlorine breakthrough readings. All affected personnel should be advised what may happen and what to do if it does happen.
Secondly, some of the facilities will be affected more than others. Those where significant bacteria build up has occurred over the previous year or so in the delivery system between the municipal source and the facility will be the most likely to experience problems.
Thirdly, if breakthrough appears to have occurred, verify the RO permeate and concentrate streams to make sure that it is a false positive reading and not a real chlorine breakthrough. There is no other known field test to confirm this.
Fourthly, review the results with the Medical Director and DON. If a false positive is determined, the problem is the testing procedure and not chlorine breakthrough in the water. Testing the permeate stream on a regular basis (such as every hour) until the false positive is gone is suggested as a prudent procedure in light of the seriousness if true chlorine breakthrough should occur during this time.
After the initial false positive reading, the problem will most likely last for only three or four days (as long as it takes to clean out the bacteria (biofilm) from the municipal lines).
It is not necessary to rebed carbon tanks. However, fresh carbon can temporarily appear to resolve the problem as the bacteria get mechanically filtered out of the water in small pore sites in the carbon. As mentioned above, this is only a temporary remedy lasting for a very short time.

Burke A. West, P.E., CWS-VI
Nelson Environmental Technologies, Inc.

2001 October 16