The fundamentals of ultraviolet (UV) light from the nature of light, light sources to targets, light absorption and transmission and UV reactor design are presented along with recommended terms and definitions.
The fundamentals of UV light and its absorption are the same for any medium: air, water or surface. In this article, I will assume that the reader has limited knowledge of UV; thus terms and definitions will be carefully defined.1 It is important to understand these fundamental terms to prevent errors, misinterpretations and misunderstandings.
Light—A particle or a wave?
Light or electromagnetic radiation encompasses wavelengths that span at least 15 orders of magnitude from gamma rays to radio waves. In the 19th century, light was considered to have only “wave-like” properties; however, in the late 19th and early 20th centuries it became clear from the work of Planck, Einstein and others that light also has “particle-like” properties. This characteristic is most apparent in the “photoelectric effect”, where Einstein was able to explain the results in terms of the concept of light as a stream of particles called “photons”. Earlier, Planck was able to explain the properties of blackbody radiation by assuming that light is composed of discrete particles with an energy inversely proportional to the wavelength. These two characteristics of light are linked together in the famous Planck Law of Radiation as shown in the following equations:
[1a] u = hν = hc/λ
[1b] U = NAhν = hcNA/λ
where u is the energy (J) of one photon, ν is the frequency (Hz = s-1), λ is the wavelength (m), c is the speed of light (2.9979 x 108 m s-1) in a vacuum, h is the Planck constant (6.6261 x 10-34 J s), NA is the Avogadro number (6.02214 x 1023 mol-1) and U is the energy of one mole or einstein of photons. The units here have been given in the standard SI forms; however, for applications in ultraviolet light and photochemistry, λ is usually given in nanometers (nm), with appropriate numerical factors to make the left-hand side of the equations come out to joules (J).
Laws of Photochemistry
Photochemical reactions are unique in that they are driven by light absorption. For example, UV disinfection is a process that is initiated by the absorption of UV photons by nucleic acid bases in the DNA of bacteria and protozoa and by either DNA or RNA in viruses. There are at least three Laws of Photochemistry2 that apply:
First Law of Photochemistry
Only the light that is absorbed by a molecule can be effective in producing photochemical change in the molecule.
If light (i.e., a stream of photons) is not absorbed as it passes through a medium, nothing can happen and no photochemical reaction can be induced. This Law is sometimes called the Grotthus-Draper Law after the works of Grotthus in 1817 and Draper in 1843. An illustrative example is the hydrogen peroxide (H2O2) molecule. Photons with wavelengths out to 560 nm have enough energy to dissociate the O–O bond in H2O2; however, no photochemical reaction occurs until light below about 300 nm is absorbed. This is because H2O2 does not absorb light above 300 nm.
Second Law of Photochemistry
Each molecule taking part in a chemical reaction caused by light absorbs one quantum of radiation (photon), which causes the reaction.
This is a consequence of the particle nature of light. This is sometimes called the reciprocity rule. It means that the photochemical yield is dependent only on the number of photons absorbed. This is sometimes called the Stark-Einstein Law after the works of Stark and of Einstein around 1912. Thus the amount of product formed or the reagent consumed will be independent of the fluence rate as long as the fluence (product of the fluence rate and the exposure time) is constant. Note that not all molecules that absorb a photon necessarily proceed to a photochemical reaction. The fraction of stimulated molecules that do react is called the quantum yield. There is an exception to this law at very high light levels (e.g., such as in a powerful laser beam), where multi-photon absorption can take place.
Third Law of Photochemistry3
The energy of an absorbed photon must be equal to or greater than the weakest bond in the molecule.
This is a consequence of the Law of Conservation of Energy. A chemical reaction generally requires one or more bond ruptures, so if the energy of the absorbed photon is less than the energy of the weakest bond, no photochemical reaction is possible. An example is NO2, a brown gas found in “photochemical smog”. NO2 absorbs light out to about 550 nm; however, only light absorbed below 395 nm has enough energy to dissociate the N–O bond. Thus light absorbed above 395 nm can only be converted to heat.
Spectral wavelength ranges of interest
shows that the range of ultraviolet is from 100-400 nm [a nanometer (nm) is 10–9 meters (m)]. “Ultraviolet” means beyond the “violet” limit (400 nm) of the visible range that extends out to 700 nm. The ultraviolet range is divided up into four sub-bands:
This sub-band extends from 315-400 nm. Light in this range is absorbed by the skin and leads largely to “sun tanning”.
This sub-band extends from 280-315 nm. Light in this range is also absorbed by the skin but leads largely to “sun burning”.
This sub-band extends from 200-280 nm. Light in this range is absorbed by DNA in the skin and is the primary cause of skin cancer. This range also is absorbed by DNA and RNA in microorganisms and leads to their inactivation by inhibiting the ability of these organisms to replicate.
This sub-band extends from 100-200 nm. It is called the “vacuum UV” since UV light in this range is strongly absorbed by water or oxygen in the air. For example, a low pressure UV lamp with a very pure quartz sleeve emits at 185 nm. This light is absorbed in a few cm by oxygen in the air and leads to the generation of ozone (O3) (See on the following page).
Some terms and definitions
The Photochemistry Commission of the International Union of Pure and Applied Chemistry (IUPAC) has developed a recommended set of terms and definitions for ultraviolet disinfection and photochemical applications (Braslavsky, 2007). In this paper, I divide the relevant definitions into those related to the light source and those related to a target being illuminated.
Radiant Power (PΦ) [W]
This is total radiant power emitted in all directions from a light source.
Radiant Energy (Q) [J]
The total radiant energy emitted in all directions from a light source is the integral of the radiant power over time.
Radiant Emittance (M) [W m–2]
The radiant emittance is the radiant power emitted in all forward directions from a tiny area (dA) on the surface of the source divided by that area. The radiant emittance is a measure of the “brightness” of a source.
Radiant Intensity (I) [W sr–1]
The radiant intensity is the power emitted outward from a source along a given direction about a tiny cone of solid angle dΩ steradians. The radiant intensity, in a non-absorbing medium, is independent of the distance from the source.
Irradiance (E) [W m–2]
The irradiance is the total radiant power of all wave lengths passing from all incident directions onto an infinitesimally small area dA, divided by dA. Often the units mW cm–2(= 10 W m–2) are used. Irradiance is measured by a radiometer and the term is appropriate for any situation where a surface is being irradiated (e.g., in UV curing).
Fluence Rate (Eo) [W m–2]
The fluence rate is the total radiant power of all wavelengths passing from all incident directions onto an infinitesimally small sphere of cross-sectional area dA, divided by dA. Often the units mW cm–2 (= 10 W m–2) are used. This is the appropriate term for UV disinfection because a microorganism in air or water can receive UV photons from many different directions, particularly in situations where there are several UV lamps. The fluence rate is the sum of the irradiance contributions from each of the UV lamps.
Fluence rate and irradiance (see Figure 2) are often confused and misused. Often the term UV intensity is used (as in the recently released US EPA UV Disinfection Guidance Manual [US EPA, 2003]). However, the term “UV intensity” does not distinguish between “fluence rate” and “irradiance”; hence, its use is not encouraged, unless one wishes to express a qualitative assessment. For example, one can say that the UV intensity from a medium pressure UV lamp is much greater than that from a low pressure UV lamp at the same distance (See Figure 2).
Fluence (F) [J m–2]
The fluence is the total radiant energy of all wavelengths passing from all incident directions onto an infinitesimally small sphere of cross-sectional area dA, divided by dA. Often the units mJ cm–2 (= 10 J m–2) are used. The fluence is the time integral of the fluence rate. If the fluence rate is constant in time, the fluence (J m–2) is the product of the fluence rate (W m–2) and the exposure time(s). In much of the current literature, the term UV dose is used for fluence. “UV dose” is not an appropriate term because the word “dose” implies complete absorption, as in UV exposure of skin. However, a typical microorganism absorbs less than one percent of the incident UV photons, since it is so small. Nevertheless, the term “UV dose” is widely used, particularly in North America. Perhaps the reason is that engineers find the term “UV dose” more intuitive than “fluence”.
Absorption and transmission
When light passes through an absorbing medium, it is apportioned into three parts: the fraction absorbed, the fraction transmitted and the fraction scattered. In most cases (where the turbidity is low), scattering can be neglected, as we will do here.
Beer-Lambert Law, Transmittance, Absorption Coefficient and Absorbance
Consider Figure 3, where a light beam of wavelength λ enters with an irradiance Eo into an absorbing medium with a path length l cm and emerges with an irradiance El. These two irradiances are connected by the Beer-Lambert Law:
T = — = 10–A = 10–al
log (T) = log — = –A = –al
where T is the transmittance, a [cm–1] is the (decadic) absorption coefficient,4 A is the absorbance (unitless) and l is the path length [cm]. These can also be based on a meter (as they are in most parts of the world except North America). Hence, the absorption coefficient is then expressed in units of m–1. Note that 1 cm–1 = 100 m–1 (See Figure 3).
Note that equations 2 apply only for monochromatic beams with a narrow distribution about a central wavelength λ.
The absorbance is a very important quantity because it is directly proportional to concentrations of absorbing components, that is:
 A = Σ εicil
where εi is the molar absorption coefficient (Mi cmi) of component i and ci is the concentration (M = mol Li) of component i.
A254 = UV absorbance at specified wavelength, based on one cm path length (unitless; absorption as measured by Standard Method 5910B).
Often the terms absorbance and absorption coefficient are confused. For example, the symbol A254 is often called the absorbance at 254 nm when it is implied that the path length is one cm. It would be better to use the symbol a254 and call it an absorption coefficient (units cm–1); then it would be clear that the path length must be one cm.
It is important to specify the path length for the transmittance, since without that specification, the transmittance is undefined. Sometimes the path length (in mm) is appended as a subscript, for example, T10 means the transmittance for a path length of 10 mm (one cm).
Factors affecting the performance of UV reactors
A UV reactor is a defined space that contains UV lamps. If used for water treatment, the UV reactor can either be an enclosed space or be an open-channel, where the top water surface is open to the air. UV reactors for air treatment are almost exclusively enclosed space reactors.
The performance of UV reactors can be expressed either as the fluence (UV dose) delivered or as the ratio of the effluent to influent concentration of a contaminant. The performance depends on several factors:
1. The absorption coefficient or transmittance of the medium (e.g., air or water)—this is probably the most important factor. Generally, as the absorption coefficient (cm–1 or m–1) increases (transmittance decreases) the fluence (UV dose) that a reactor can deliver at a fixed flow rate decreases. Figure 4 (taken from Bolton et al. 2001) illustrates how the performance deteriorates as the percent T10 drops for an annular UV reactor. This curve should apply equally well to more complex reactors.
The percent T of the medium also is important in UV reactor design. Figure 5 (taken from Bolton et al. 2001) illustrates how quickly the UV fluence rate (irradiance) falls off with distance as a function of the percent T of the medium. For drinking water rectors, where the percent T10 is often >90 percent, the penetration depth is quite large (8–12 cm), so reactors should be designed so that lamps are relatively far apart and the walls are not too close to the outer lamps. By contrast, wastewater has percent T10 values that are 40–60 percent or less. Here penetration depths are only one to two cm. Thus reactors should be designed with lamps relatively close to each other and the walls close to the outer lamps.
2. Number and power of the UV lamps—generally, the fluence (UV dose) delivered increases linearly with the power applied to the UV lamps, although this is not true when, for example, a low pressure lamp is replaced by a medium pressure lamp because the UVC efficiency of the latter is less than half that of the former.
3. Flow Rate—generally, the fluence (UV dose) decreases as the flow rate increases because the residence time in the reactor is inversely proportional to the flow rate. However, the dependence is often non-linear because at high flow rates, the mixing efficiency improves.
4. Mixing efficiency—this is defined as the ratio of the actual fluence (UV dose) delivered at a given flow rate to the maximum theoretical fluence (UV dose) assuming perfect radial mixing as the air or water passes through the reactor. The latter can be determined by calculating the volume average fluence rate (using a suitable mathematical model) and multiplying by the residence time in seconds.
5. Reflection—UV light (200–300 nm) reflected from the reactor walls back into the reactor can significantly improve performance, since not all this UV light is lost at the walls. Aluminum has the best reflection coefficient (>95 percent) whereas it is only about 25 percent for stainless steel and virtually zero for wood.
This paper has attempted to lay out the fundamentals of ultraviolet light in its application to UV reactors for air and water treatment. It is important to use properly-defined terms and units, so that readers can have a clear view of the concepts being put forward.
A good source of UV references is the reference list available in the Member Zone of the International Ultraviolet Association (IUVA) (www.iuva.org).
- Braslavsky, S.E., Glossary of Terms use in Photochemistry, 3rd Ed. Pure Applied Chemistry, 79 (3): 293-465 (2007).
- Bolton, J.R. (2001) Ultraviolet Applications Handbook, 2nd Ed., Bolton Photosciences Inc., 628 Cheriton Cres., NW, Edmonton, AB, Canada T6R 2M5.
Bolton, J.R., Stefan, M.I., Cushing, R.S. and Mackey, E. 2001. The Importance of Water Absorbance/Transmittance on the Efficiency of Ultraviolet Disinfection Reactors, Proc. First International Congress on Ultraviolet Technologies, June 2001, Washington, DC. CD/ROM published by the International Ultraviolet Association, P.O. Box 1110, Ayr, ON, Canada N0B 1E0.
- Calvert, J.G. and Pitts, J.N. (1966) Photochemistry, Wiley, New York.
- USEPA (2006). Ultraviolet Disinfection Guidance Manual for the Final Long Term 2 Enhanced Surface Water Treatment Rule. Available on the web at http://www.epa. gov/safewater/disinfection/lt2/comjpliance. html
- Wayne, R.P. (1988) Principles and Applications of Photochemistry, Oxford University Press, Oxford, UK.
James R. Bolton is Executive Director of the International Ultraviolet Association. He can be reached at 628 Cheriton Cres., NW, Edmonton, AB, Canada T6R 2M5. Email: email@example.com