Ultraviolet (UV) technology was originally used to ensure the adequate disinfection of municipal towns mains water. Since its introduction over 40 years’ ago, it is now applied globally for disinfection, TOC (total organic carbon) reduction, destruction of Ozone and Chloramines plus de-chlorination of process water in many different industries, including pharmaceutical manufacturing. Water is the largest volume material used within pharmaceutical processes and, driven by more stringent standards, increasingly sophisticated process barriers and disinfection techniques have been adopted. The United States Pharmacopoeia 24th edition (USP 24) defines the quality standard to which water used in this pharmaceutical manufacturing needs to be treated.
Several of the process stages in pharmaceutical production can themselves also cause microbial contamination and UV can therefore be used as an effective barrier to ensure that discrete process stages do not compromise quality standards.
Typical installations include UV for disinfection after carbon filters or before RO and UV for disinfection and TOC reduction in the polishing loop. A correctly sized UV disinfection system installed downstream of the carbon beds or directly upstream of the RO unit will eliminate at least 99.9% of bacteria present in the inlet water.
Disinfection with UV technology
UV disinfection systems are generally split into two distinct types: low pressure (LP) and medium pressure (MP). LP systems have a monochromatic UV output (limited to a single wavelength at 254nm), whereas MP systems have a polychromatic UV output (with an output between 240-310nm).
In essence, UV works by fusing Adenine and Thiamine molecules within a micro-organism’s DNA, rendering it unable to replicate. The micro-organisms are thus destroyed without the use of chemicals. Whilst 254nm is an effective wavelength for disinfection, it is generally accepted that DNA absorbs UV most effectively at 265nm, a wavelength that MP lamps produce in abundance. Understanding these differences is fundamental to the design of efficient and effective UV disinfection equipment. Generally speaking, LP systems are best used on small, intermittent flow applications with MP technology lending itself to higher flowrates.
For LP systems to be effective at destroying DNA, a large number of UV lamps (the mercury-filled tubes that actually produce the UV) are required. This has obvious cost and maintenance implications for the operator. All UV lamps have set lifespans and need replacing after a certain number of hours (normally several thousand). The more lamps there are, the greater the likelihood that the system will need to be stopped to replace those that fail. UV monitoring is also more difficult with a large number of lamps – UV monitors are located on the wall of the UV chamber, so if a lamp located away from the chamber wall fails, it may not be detected. In addition, a large number of lamps impedes fluid flow through the UV chamber, resulting in pressure headloss and higher pumping costs.
In addition to the above points, LP lamps can also be difficult to clean. For instance, Iron in solution will often form deposits on the quartz sleeve surrounding the lamp. This affects lamp efficiency and must be kept to a minimum. To overcome this problem MP systems utilise a mechanical wiper which passes back and forth along the length of the quartz sleeve, keeping it clear of iron deposits or other debris. LP lamp systems typically rely on chemical cleaning. This usually requires the systems to be completely stripped and the sleeves hand-cleaned – a time-consuming process.
For these reasons many operators are switching to the newer, more efficient MP technology for higher flowrates. With a wider – and more powerful – UV output than LP lamps, far fewer MP lamps are required for the same level of disinfection. Headloss is significantly reduced and monitoring is far more effective – an essential requirement in pharmaceutical manufacturing. They are not temperature sensitive and have rapid start-up times, making them more suitable for complex process applications such as those found in the pharmaceutical industries. Maintenance costs are also reduced as there are physically fewer lamps to replace. Typically, 10-12 LP lamps are required to produce the same UV output as one MP lamp.
LP products are usually used in low flow applications, often installed in the heart of water treatment skids. Small, single lamp units can be effectively monitored and economically used to treat flows up to around 50m3/h.
Installing UV systems
UV units can be installed at various points along an ultrapure water system. Installation or retrofitting to existing pipework and vessels is relatively straightforward, requiring minimum disruption and site preparation. Depending on the level of use, the only routine maintenance required is changing the arc-tubes every 9-12 months, a simple procedure that can be carried out by on-site personnel. Once installed, the processing plant can be kept operational 24 hours a day, without the necessity of shutting down the system for routine sanitation and sterilisation.
UV dose is computed by using three independent variables:
UV dose (fluence) = Lamp Intensity X Residence Time Distribution X Water Transmittance
To ensure that the UV dose (lamp output) is effectively measured, each of the process variables needs to be measured. Many UV monitors have adjustable potentiometers to allow simple re-calibration. This does not make the measurement relative, nor absolute. The monitor camera should be sealed, and calibrated against a traceable norm. An audit trail should be provided with each lamp and monitor to ensure that the lamp output, measured in watts/cm-2, of germicidal UV is measured, not guessed. The same is true of the monitor response, measured in mW/cm-2. No in-field adjustment to the monitor camera should be possible, and normally these cameras should be returned to the manufacturer for re-calibration in accordance with the audit trail.
Each lamp should have a unique serial number, together with a certificate of spectral conformity. It is standard practice to be able to measure, not infer, UV dose expressed in mJ/cm-2 with a dedicated monitor camera for each lamp. Those who seek to apply Good Manufacturing Practice are able to data-log the UV dose received by the water, and the validation of the process is completed with an event stamp of any UV fault, showing date and time, and providing a permanent record of the recorded fault.
Recent research by Hanovia has shown that short wavelengths (below 200nm) are highly effective at breaking down organic molecules present in water, especially low molecular weight contaminants. Experiments carried out by the company using a PFW (purified water) loop showed that below 200nm UV works in two ways: the first method is by direct photolysis, when energy from the UV actually breaks down chemical bonds within the organics, the second method is by the photolysis of water molecules, splitting them to create charged OH- radicals, which also attack the organics.
It was also discovered that increasing the power input to the UV system was actually found to be detrimental to effective treatment, as higher wattage per unit surface area increases the temperature of the quartz sleeve of the arc tube, shortening its life and inhibiting the shorter wavelength output. Higher power input causes the internal pressure of the arc tube to rise as well, which also cuts off short wavelength UV.
Traditionally, MP lamps have been used for TOC reduction, but their output below 200nm is relatively low, limiting their effectiveness. Hanovia has therefore developed a new type UV lamp designed specifically for TOC reduction. Called SuperTOC, this lamp has a sub-200nm output between two and three times that of any previous MP lamp. It is designed to operate at lower temperatures than conventional MP lamps, increasing lamp life and optimising UV output below 200nm, where it has the greatest potential for reducing TOC.
The make-up water to many pharmaceutical plants is derived from municipal water supplies and, for over 50 years, free chlorine has been widely used for residual disinfection. When chlorine is injected into waters with naturally occurring humic acids, fulvic acids and other naturally occurring materials, trihalomethane (THM) compounds are formed. Since some THM’s have been demonstrated to be cancer-causing to laboratory animals in relatively low concentrations, the U.S. environmental protection Agency has set their maximum contaminant level in primary drinking water to be 100 parts per billion (ppb) since 1979.
In addition, because of its properties, chlorine can damage delicate process equipment like reverse osmosis (RO) membranes and deionization (DI) resin units and must be removed once it has performed its disinfection function.
To date, the two most commonly used methods of chlorine removal have been granular activated carbon (GAC) filters or the addition of neutralising chemicals such as sodium bisulphite and sodium metabisulphite. Both of these methods have their advantages, but they also have a number of significant drawbacks. GAC filters, because of their porous structure and nutrient-rich environment, can become a breeding ground for bacteria. Dechlorination chemicals such as sodium bisulphite, which are usually injected just in front of RO membranes, can also act as incubators for bacteria, causing biofouling of the membranes. In addition, these chemicals are hazardous to handle and there is a danger of over- or under-dosing due to human error.
Medium pressure UV is now becoming increasingly popular as an effective alternative method of dechlorination. It has none of the drawbacks of GAC or neutralising chemicals, while effectively reducing both free chlorine and combined chlorine compounds (chloramines) into easily removed by-products.
Between the wavelengths 180nm to 400nm UV light produces photochemical reactions which dissociate free chlorine to form hydrochloric acid. The peak wavelengths for dissociation of free chlorine range from 180 nm to 200 nm, while the peak wavelengths for dissociation of combined chlorine (mono-, di-, and tri-chloramine) range from 245nm to 365nm. Up to 5ppm of chloramines can be successfully destroyed in a single pass through a UV reactor and up to 15ppm of free chlorine can be removed.
The UV dosage required for dechlorination depends on total chlorine level, ratio of free vs. combined chlorine, background level of organics and target reduction concentrations. The usual dose for removal of free chlorine is 15 to 30 times higher than the normal disinfection dose of 30,000 microWatt-seconds per centimetre squared (mW-s/cm2). Additional important benefits of using UV dechlorination are high levels of UV disinfection, total organic carbon (TOC) destruction and improved overall water quality at point-of-use.
UV is a key process tool that can ensure purified water loops operate at the highest levels of microbiological integrity. Its benefits are many: installation is easy, requiring little disruption to the plant, maintenance is simple and can be carried out by on-site personnel, as a non-chemical method of treatment, there is no possibility of a detrimental effect on product stability and products are also free from unwanted residues, colours and odours.
Independent audit trails now allow UV dose measurements to be accurately measured, not inferred, with the intensity calibrated against an absolute standard. In addition, data-logging ensures that compliance can be measured and demonstrated, not simply guessed. It is also being successfully utilised for TOC reduction, dechlorination and dechloramination by some of the world’s leading pharmaceutical manufacturers.