Ultraviolet radiation imaging

Imaging > Imaging Techniques > Ultraviolet Radiation Imaging

In progress: Seeking additional comments and images to develop this section

Please be patient, this part of the site is under development. We are starting to build out the Imaging Wiki.

Description of technique ... Include a short paragraph defining and introducing the technique.

• Applications info and touching on functions of Ultraviolet imaging

Wiki Team Lead: Yosi Pozeilov, Jennifer McGlinchey Sexton
Wiki Editor: Gwenanne Edwards, Chantal Stein "your name could be here"
Wiki Contributors: "your name could be here"

A note on terminology: Terminology is tricky and there is not yet standardized terminology for conservation imaging techniques. For the IWG Imaging Wiki, we recommend the use of luminescence instead of fluorescence. Luminescence is the more general term, while fluorescence requires knowing the lifetime of the emission. The use of luminescence is in alignment with the CHARISMA imaging user manual (Dyer, Verri, and Cupitt 2013) and “UV-Vis Luminescence: Imaging Techniques” (Picollo, Stols-Witlox, and Fuster López 2019). For additional references about this terminology decision, see Verri et al. (2008), Bacci (2019), and Picollo, Stols-Witlox, and Fuster López (2019, p. 13-15).

Introduction

What is UV radiation?

Ultraviolet radiation refers to the band of the electromagnetic spectrum consisting of shorter wavelengths than visible light, ranging from 10-400 nm. Ultraviolet radiation cannot be seen by the human eye and is often termed "invisible," as are all other electromagnetic radiations except those in the narrow visible range. Silicon solid-state imaging sensors (when bare or unfiltered) have limited sensitivity to the longer wavelengths of ultraviolet radiation just below 400 nm. Filters that absorb all visible light but allow passage of this  truncated section of ultraviolet radiation make it possible to record the ultraviolet radiation photographically. This is the basis of ultraviolet imaging.

Image from https://www.istockphoto.com/photos/uv-spectrum

Regions of Ultraviolet Radiation

The ultraviolet spectrum is divided into four main bands: longwave UVA, middle wave UVB,  shortwave UVC, and vacuum ultraviolet (UVD) (Warda et al. 2017, Davies 2018). The first three bands can be used for imaging, but the fourth cannot.

Longwave Ultraviolet Radiation (UV-A)

Longwave ultraviolet radiation (UV-A) extends from about 320-400 nm. This band is sometimes called "near ultraviolet," since it is the closest to the visible range. This range of ultraviolet radiation is transmitted by the regular optical glass used to manufacture most photographic lenses, therefore, it is the most commonly used range in ultraviolet photography.

Middle Wave Ultraviolet Radiation (UV-B)

Middle wave ultraviolet radiation (UV-B) extends from about 280-320 nm. Middle wave ultraviolet radiation is not transmitted by regular photographic lenses. A lens made of quartz will transmit these rays; however, quartz lenses are costly and therefore not as readily accessible.

Shortwave Ultraviolet Radiation (UV-C)

Shortwave ultraviolet radiation (UV-C) extends from about 200-280 nm. This band is sometimes called "far ultraviolet" since it is farthest from the visible range. A quartz lens will still transmit most wavelengths of shortwave ultraviolet.

What is UV imaging?

Ultraviolet imaging can be divided into two categories: reflectance and luminescence imaging. The division emphasizes that these are two different types of imaging, the former recording the reflected ultraviolet radiation, and the latter recording the emitted radiation or photo-induced luminescence from ultraviolet irradiation of the object (Dyer, Verri, and Cupitt 2013). This distinction is important since ultraviolet imaging, as a generic term, has often been colloquially used to denote luminescence imaging. Here, an effort is made to distinguish between them clearly.

Reflected ultraviolet (RUV) imaging records ultraviolet radiation that is reflected from, absorbed by, or transmitted by the materials that make up an object.  RUV requires a camera sensitive to UV radiation, filters on the camera to limit the transmission of wavelengths to only UV, and a UV radiation source. RUV can be used to differentiate materials and examine surfaces (Warda et al. 2017; Dyer, Verri and Cupitt 2013). Warda et al. (2017, p. 158) note the use of RUV to differentiate materials to identify past conservation treatments, visualize faded or obscured features (e.g., faded iron gall ink inscriptions), and recording the surface topography of an object since UV radiation does not penetrate surfaces deeply. Dyer et al. (2018) include the use of RUV to detect patterns, staining and surface coatings on textiles.

Ultraviolet-induced visible luminescence (UVL) imaging records the visible luminescence resulting from radiation excitation by materials within an object. Luminescence is  a  phenomenon in which radiation is absorbed by a chemical molecule at a specific wavelength (the excitation wavelength) and then emitted by that molecule at a longer wavelength (the emission wavelength). Fluorescence and phosphorescence are two types of luminescence. The main difference between the two is the latent period of emission: for fluorescence the emission period is very short, measured in milliseconds, whereas for phosphorescence the emission period can be much longer, from seconds to minutes. For additional information about the luminescence see Bacci (2019). Luminescence phenomena is further discussed below. Though various types of radiation can cause luminescence in different regions of the electromagnetic spectrum, within the cultural heritage field ultraviolet radiation is most widely used to excite visible luminescence  which can be observed by the unaided eye (and is used for visual examination) or recorded with UVL imaging.  UVL is used to visualize, spatially locate, and differentiate materials, establish the condition, and reveal past treatments (Mairinger 2000; Warda et al., 2017, pp.149-150; Webb 2019). Specific examples of materials for each of these areas are included in Webb (2019, pp. 38-43). UVL can be used to map the spatial distribution of luminescent materials on an object (Dyer, Verri, Cupitt 2013).

In addition to UVL, ultraviolet radiation can be used for ultraviolet-induced infrared luminescence (UVIL) which can be used for mapping the spatial distribution of some pigments including red and yellow cadmium and the differentiation of rutile and anatase titanium white (Keller et al., 2019).

History of UV Radiation Use and Museum Applications

The term "fluorescence" comes from the mineral fluorite, which glows faintly blue under ultraviolet light from sunlight (Marfunin 1979:143).

One of the earliest mentions of luminescence dates back to the 17th century, when Vincenzio Cascariola created phosphorescent barium sulfide (Bologna Phosphorus) by burning barite. His findings, published by La Galla after learning of the substance from Galileo, sparked early interest in the phenomenon (Marfunin 1979:143).

Systematic studies of luminescence began in the mid-1800s. In 1810, Johann Wolfgang von Goethe noted that some minerals fluoresce. Early researchers, including Sir David Brewster and Sir John Herschel, identified luminescence as variations of known light properties, such as diffusion and dispersion (Robbins 1983:3-13).

Sir George Stokes, building on Goethe’s work, observed fluorite glowing under ultraviolet light and coined the term "fluorescence" based on his findings (Dake and De Ment 1941:1-7; Radley and Grant 1954:4-10). Advances in ultraviolet sources began in 1903 with the iron arc lamp, leading to the first public display of fluorescent minerals at the British Museum of Natural History (Radley and Grant 1954:11-13).

In the 1920s and 1930s, new ultraviolet radiation sources emerged. The argon bulb, which produced low-intensity UV light, was followed by the more effective but costly Nico lamp. Dr. Robert Wood then developed a glass filter that passed only ultraviolet light, leading to mercury vapor lamps known as Wood’s lights. These lamps became common in conservation labs before ultraviolet LEDs. Wood’s nickel-plated filter emitted UV-A light (320-400 nm), peaking at 365 nm. Shortwave UV-C lamps (200-280 nm, peaking at 254 nm) are also used in non-destructive artwork examination (Radley and Grant 1954:11-13).

Art historians, curators, and conservators have long relied on UV-induced luminescence to reveal materials like glues, varnishes, and resins, helping detect hidden repairs on artwork. The invention of Wood’s light in the late 1920s led to the widespread use of UV for art examination, first by curators and later by conservators, due to its non-destructive diagnostic capabilities (Radley and Grant 1954:11-13).

In 1931, the Metropolitan Museum of Art published James Rorimer’s Ultra-Violet Rays and Their Use in the Examination of Works of Art, one of the earliest references to UV use in museums. Rorimer advocated for UV imaging to record luminescence and ensure reproducibility, though he mentioned few standards (Rorimer 1931).

Another key work, Fluorescence Analysis in Ultra-Violet Light (Radley and Grant 1933), discusses the use of UV to distinguish authentic objects from forgeries and enhance surface examination.

Today, UV radiation remains a vital tool for identifying surface inconsistencies, such as inpainting or fills. It is particularly useful in diagnosing variations in paper, textiles, varnishes, and ceramics. UV emitters are affordable, easy to use, and provide immediate information, making them a valuable, non-destructive tool for conservators.

Radiation Sources

There are many natural and artificial sources of ultraviolet (UV) radiation that can be used for UV imaging, with the choice depending on factors like availability, cost, area coverage, intensity, and specific emission wavelength (Davies 2018, pp. 43-50; Warda et al. 2017, pp. 152-153). Verri (2019) provides a more technical discussion on UV sources, including the effects of parasitic and stray radiation.

Sunlight

Sunlight is a natural source of longwave UV radiation and can be used for UV reflectance (RUV) photography under dry, clear conditions. While longwave UV passes through the atmosphere, shortwave UV is absorbed or scattered. For RUV imaging, a filter is placed over the camera lens to block visible light and transmit longwave UV. However, sunlight is impractical for UV luminescence (UVL) imaging, as this technique requires a completely dark environment, which is hard to achieve outdoors. Sunlight is also inconsistent due to atmospheric changes and the movement of the sun.

Mercury Vapor Lamps

For controlled imaging, mercury vapor lamps are widely used. Low-pressure mercury lamps, often called "black light" tubes, emit longwave UV radiation through ionized mercury vapor and are widely available, inexpensive, and capable of illuminating large areas. However, these lamps often emit some visible blue light, which can interfere with imaging and may require additional filters to block unwanted wavelengths. High-pressure mercury lamps, on the other hand, emit higher UV output across longwave, middle wave, and shortwave regions, making them useful for small-area imaging and photomicrography. These lamps provide intense UV radiation and are ideal for applications requiring strong excitation, but they are bulkier, more expensive, and produce more heat and ozone.

Arc Lamps

Arc lamps are another option, commonly used for generating medium and longwave UV radiation. Xenon arc lamps, in particular, produce a continuous spectrum from UV to infrared and are mostly used in visible light photography, though they emit useful UV for imaging. Electronic flash lamps also emit UV radiation, especially in the longwave range, and are useful for UV reflectance imaging. For luminescence imaging, however, their short flash duration can be a drawback. This can be addressed by using a multi-pop technique, where multiple flashes build up luminescence exposure while the camera shutter remains open in a dark environment (Verri and Saunders 2014).

UV-LEDs

In recent years, UV-LEDs have become a popular source of UV radiation due to their efficiency, compact size, and stability. UV-LEDs emit defined peaks in UV wavelengths, ranging from longwave (UV-A, 350-420 nm) to medium wave (UV-B, 300-350 nm), and they do not require warm-up periods or heavy filtering like mercury vapor lamps. They are used in various fields, including forensics, art conservation, and environmental applications. UV-A LEDs are the most common and are ideal for applications requiring 365 nm excitation, such as museum and archive imaging. While medium wave UV-LEDs are still in development, they hold potential for uses in curing materials, biomedical applications, and DNA analysis. Shortwave UV-LEDs (UV-C, 250-300 nm) are also being developed for air and water purification systems, though they are not yet widely available. UV-LEDs offer significant advantages over traditional UV sources, particularly due to their environmental safety and reduced energy consumption, making them an increasingly popular choice for UV imaging.

Health and Safety

Ultraviolet rays have more energy than visible light but not as much as x-rays. Higher energy ultraviolet rays are a form of ionizing radiation. This means they have enough energy to remove an electron from (ionize) an atom or molecule. Ionizing radiation can damage the DNA in cells, which in turn may lead to cancer. But even the highest energy ultraviolet rays don’t have enough energy to penetrate deeply into the body, so their main effect is on the skin and eyes. The different bands of ultraviolet radiation can cause various adverse health effects.

UV-A rays have the least amount of energy. These rays can still cause skin cells to age and can cause some indirect damage to cells’ DNA. UV-A rays are mainly linked to long-term minor skin damage such as wrinkles, but they are also thought to play a role in some skin cancers.

UV-B rays have slightly more energy than UV-A rays. They can damage the DNA in skin cells directly, and they are the main rays that cause sunburns. They are also thought to cause most skin cancers.

UV-C rays have the most energy. In nature, UV-C radiation reacts with ozone high in our atmosphere and is prevented from reaching the ground level, so it is not normally a risk factor for skin cancer. But, as mentioned before and as discussed further below, UV-C rays can also come from certain man-made sources, such as arc welding torches, mercury lamps, and UV sanitizing bulbs used to kill bacteria and other germs. These radiation sources offer the most potential for damage based on exposure time to the human body, from immediate “sunburn” (akin to a second or third degree burn) to blindness and deep damage to gene information at a DNA level.

Working with ultraviolet (UV) light in the lab can pose risks for both conservators and the objects being studied. UV radiation, especially in the UVB range, can accelerate skin aging and increase the risk of skin cancer. Although outdoor UV exposure is typically more intense, damage from UV radiation is cumulative, so it’s important to take protective measures even indoors. Wearing UV-absorbing glasses is essential when working with any UV source, as prolonged exposure can lead to cataracts, and acute exposure—particularly to UVB and UVC—can severely damage the cornea. Additionally, when working with UVB or UVC, it’s important to cover exposed skin to minimize the risk of burns. While objects are exposed to higher levels of UV radiation during examination compared to photography (which only requires a few seconds per image), precautions should still be taken to limit exposure (Warda et al. (2017, p. 151-152).

Precautions

Ultraviolet Safety Recommendations:

• Always wear UV-protective goggles and minimize skin exposure when working with UV sources.
• Use a task light during fluorescence examinations to reduce strain on your eyes and ensure a more effective comparison between visible and UV light views of the object. Allow time for your eyes to adjust to the dark for better fluorescence visibility.
• Be mindful of the intensity of UV radiation, especially when lamps are close to the object, and keep exposure times brief. Also, high-pressure UV lamps can generate significant heat, which can damage both the object and the conservator.
• Whenever possible, keep lamps directed away from the object when not actively in use.

All ultraviolet emitting devices should have warning labels similar to the one shown below.

Techniques

Reflected ultraviolet (RUV) imaging

Ultraviolet reflected imaging records information that cannot be seen with the human eye. Ultraviolet reflected imaging is based on the premise that various elements of an object will reflect or absorb ultraviolet radiation to different degrees. Some materials will absorb ultraviolet radiation while others will reflect it, and some materials have partial absorption and partial reflection. These effects can be recorded photographically and used to differentiate materials. Visible light and infrared imaging operate on the same principles of absorption and reflection, with the exception that visible light imaging provides information about hue and chroma in addition to lightness. For ultraviolet and infrared imaging, the information is expressed only in lightness values and yields a grayscale image.

UVR imaging can be used to differentiate between materials and examine the surface of an object (Warda et al. 2017, p. 157-160). The shorter wavelengths of ultraviolet radiation tend to be absorbed at the surface, and RUV generally provides information about  the surface layer of the object being imaged. The technique is particularly useful for looking at the superficial distribution of materials (e.g., varnishes and coatings) (Dyer, Verri, and Cupitt 2013). Some white pigments can be differentiated (e.g., zinc and titanium white are strong absorbs while lead white is reflective) (Warda et al. 2017, p. 158; Keller et al. 2019). Iron gall ink absorbs UV radiation, and UVR can provide increased visibility even when significantly faded (Warda et al. 2017, p. 158). For textile documentation and analysis, UVR can be used for detecting and recording patterns, staining, surface coatings, and past restorations (Dyer et al. 2018). UVR images are also used with reflected visible light images to process false color ultraviolet (FCUV) images. More information about FCUV can be found on the False Color Processing wiki page.  

Equipment

The three main components of an imaging system for ultraviolet imaging (both reflectance and luminescence) are (1) a recording device, or camera, (2) radiation sources, and (3) filters that can be used on both the camera and/or the radiation sources.

Cameras

RUV records images within the UV region of the electromagnetic spectrum and requires a camera that is sensitive to UV radiation. A standard digital camera is not sensitive to UV radiation without modification. The silicon sensors at the heart of consumer digital cameras are inherently sensitive from near UV, through visible, and into the infrared. These consumer digital cameras include internal filters that block UV and IR wavelengths to produce the highest quality visible, color images. Removing this internal blocking filter is known as a full-spectrum UV-VIS-IR modification, which allows the camera to be used for a range of multiband imaging techniques in addition to visible light imaging by changing the filters on the camera lens and the radiation sources. Another type of modification is a monochrome conversion where the internal blocking filter is removed in addition to the color filter array on the sensor surface that is used to produce a color image. This modification is a more intrusive and expensive conversion, and it is not as widely used for conservation applications. The monochrome modified camera cannot be used for color visible images or color UVL images.

Radiation Sources for RUV

See the above section on radiation sources. Note that sunlight can work for RUV, but not UVL.

Filters

For UV imaging (both reflectance and luminescence), there are two different types of filters used: exciter filters and barrier filters (Mairinger 2000, p. 60). The exciter filters transmit long wave ultraviolet radiation and block visible light and IR radiation. Exciter filters can be used either on UV radiation sources for UVL or on the camera lens for RUV. Barrier filters transmit visible light and block UV and IR radiation in order to record the visible luminescence for UVL imaging. Filters are manufactured by different companies (e.g., Kodak, Schott, Peca, Hoya, B+W, etc.). While there are equivalent filters between different companies, these may vary in terms of transmission and comparing images acquired with different filters and setups may not be consistent (Chen and Smith 2020).    

Regardless of the radiation source, for reflected ultraviolet imaging (RUV), a filter must be placed over the camera lens to filter out visible light and infrared radiation from the ambient environment and reflected from the object. The filter should ideally have a high transmittance of ultraviolet radiation and should block all visible light and infrared radiation. Ultraviolet transmitting filters are usually manufactured from glass containing coloring agents that control transmittance. Most types of glass transmit longwave ultraviolet radiation while absorbing medium and shortwave ultraviolet radiation; this is not problematic for ultraviolet reflected imaging since camera lenses are also made of optical glass and also only transmit longwave ultraviolet radiation.

The Kodak Wratten 18A (Wood’s Glass filter) is a glass filter with high percentage transmittance of longwave ultraviolet radiation, peaking at 365 nm, that was traditionally used for analogue RUV photography (Davies 2018, Kodak Photographic Filters Handbook 1990). As can be seen in the following figure of the transmission of the 18A filter, there is also IR transmission in addition to the UV transmission. This IR ‘leakage’ was not an issue with analog RUV photography because the film was not sensitive in the IR, but this is an issue with modified full-spectrum UV-VIS-IR cameras that have high IR sensitivity (Davies 2018, p. 40). A range of filters have been used for RUV, but the following includes some recommendations and references. The AIC Guide (Warda et al., p. 161) provides two options for filters for RUV imaging:

1. Baader Planetarium Ultraviolet Venus Filter (Baader-U filter) which transmits between 310-390 nm.

OR

1. Kodak Wratten 18A filter (or an equivalent) + X-Nite CC1 (or Schott BG38) to block the IR transmitted by the 18A or X-Nite BP1 (more effective than CC1 or BG38)

• Kodak Wratten 18A equivalents: X-Nite 330, Peca 900, B+W 403, Hoya U-330

The following is a table from Poseilov (2015, p. 106) and includes comparable filters that transmit UV (left column) and absorb UV (right column).  

Exciter Filters / UV Transmitting Filters for RUV (for use on camera, but also can be used on radiation source if necessary)
Useful Transmission range [nm]	Filter Designation
320 - 385	B+W 403
270 - 375	MaxMax X-Nite BPI
330 (85% Peak)	MaxMax X-Nite 330C
300 - 400 (365 peak)	Kodak 18A
	Peca 900
	Schott UG1
	MidOpt BP365
	Kopp 1041
250 - 480 (254 peak)	Kodak 18B
	Chance OX 7
	Kopp 9863
	Corning 9863
250 - 380	Schott UG 5

Ultraviolet transmitting filters can also be obtained from other filter manufacturers. It is suggested, however, that filter transmittance curves be obtained and examined for efficiency of both ultraviolet transmittance and visible light absorption. For ultraviolet reflected imaging, the filters should have no visible light transmittance.

When mounting the filters to the camera lens, it is imperative that all visible light be excluded, possibly with some kind of light seal around the filter. Targets and standards can be used to ensure that the imaging environment is adequate and that the right parameters are being met to perform the best imaging possible (see Targets section).

RUV Standards, Targets, and References

Repeatability and comparability are challenging for multiband imaging and this includes RUV. Discussions around repeatability and comparability are not new and can be found in references like Dyer et al. (2013) and Webb (2019). Reflectance standards can be used to ensure that no stray light (visible or IR) has entered the camera during RUV, and can also help to achieve an appropriate exposure for the image. Generally, any standard should be carefully handled to avoid hand oil contact with the surface, as this will compromise the standard.

Spectralon diffuse reflectance standards can be used for RUV. These standards are durable and chemically inert and provide a flat spectral response from UV into the IR (250-2500 nm). These are used for reflectance calibration for techniques like multispectral and hyperspectral imaging and they can be useful references for multiband imaging and techniques like UVL and RUV (as outlined in Dyer et al. 2013). By placing these standards in the same lighting conditions as the artwork, and using their known reflectance percentages (e.g., 99%, 50%, 2%), researchers can correct for lighting variations and quantify the reflectance properties of the surface. This allows for precise analysis of surface materials and condition, as well as the comparison of UV reflectance across different imaging sessions.

[Add image of Spectralon]

The Cultural Heritage Science Open Source Technical Photography and Multispectral Imaging (CHSOS TP-MSI) Calibration Card comes with the purchase of CHSOS Pigment Checker. The target includes shellac, madder lake, zinc white, and cadmium which have characteristic luminescence and can be useful for UVL imaging while the greyscale patches can be used for RUV. The CHSOS website includes exposure recommendations for the different multiband imaging techniques  (https://chsopensource.org/technical-photography-msi-calibration-card/).

[Add image of CHSOS TP-MSI Calibration Card]

It should be noted that many of these commercial options are fairly expensive, which is met by the development of bespoke/custom targets and simple, low-cost solutions.

For example, Pozeilov (2023) presented a simple and more economical option for UV reflectance that can help to assess aspects of the imaging environment (i.e., camera filtration, radiation source emission, and ambient radiation) using PTFE tape and highly reflective metal sheet.

RUV Capture Considerations

In order to photograph an object using reflected ultraviolet radiation, the following steps are necessary and best done in a completely dark environment:

1. Use an ultraviolet radiation source to irradiate the object as evenly and uniformly as possible across its surface.
2. Exclude all visible light from reaching the imaging sensor inside the camera. This can be done by placing an exciter filter over the lens that transmits only ultraviolet radiation and excludes all other wavelengths.
3. The recording device must be sensitive to ultraviolet radiation. For cultural heritage and conservation applications, these recording devices are modified cameras with the internal “hot-mirror” filter removed so that the sensor is made sensitive to the full spectrum capture capabilities of the silicon chip.

Warda et al. (2017, pp. 175-162), Pozeilov (2015, p. 108), and Dyer, Verri, and Cupitt (2013) provide recommendations for equipment, setup, and processing of UVR images.

RUV Exposure

Exposure in UV reflectance imaging is typically determined by visual inspection or through histogram analysis. Standard grayscale targets are not reliable in UV imaging since the UV reflectance of these patches differs significantly from their visible light reflectance.While the best option for exposure for UVR imaging is using a 99% Spectralon diffuse reflectance standard, these are expensive and may not be included in everyone’s imaging setup. The targets more widely used for visible light photography are not intended to be used for UVR (e.g., X-Rite ColorChecker, AIC PhD, etc.).

The AIC Guide provides exposure recommendations for RUV based on the X-Rite ColorChecker N8 patch. When capturing images in RAW format using a DSLR, open them in an image editing software and desaturate the image (set saturation to -100) to view it in black-and-white before adjusting exposure. For the Neutral 8 patch in UV imaging, adjust the RGB value to 120 (+/- 5) as a starting point, and then refine as needed to highlight key areas of interest (Warda et al. 2017, p. 162).

Ultraviolet-Induced Visible Luminescence Imaging

What is Luminescence?

The basic concept of molecular luminescence involves how substances interact with light or other forms of electromagnetic energy. When this energy hits a substance, it can either pass through or be absorbed, depending on the substance’s molecular structure and the energy’s wavelength. If the molecule absorbs energy, it stores it by increasing the motion of its electrons, either through vibration or rotation. When enough energy is absorbed, the molecule enters an excited state, which is unstable. To return to a stable state, the molecule releases this extra energy in the form of light, often in the visible spectrum. This process of emitting light is called luminescence. There are two main types of luminescence: fluorescence, where the emission happens almost immediately, and phosphorescence, where the emission is delayed.

Fluorescence

When the luminescence ceases within a very short timeframe (10^8 seconds) after the exciter radiation is removed, the phenomenon is termed fluorescence. Although fluorescence is commonly produced by excitation with ultraviolet energy, other types of radiation can also be used. The fluorescence produced will always be a lower energy than the original radiation (for example, UV radiation can produce visible light fluorescence, High energy x-rays or gamma ray radiation can produce lower energy x-ray fluorescence, visible light radiation can produce infrared fluorescence, etc.).

Many thousands of materials fluoresce. Fluorescence photography has numerous applications because it can provide information that could not otherwise be obtained by other photographic methods. The fact alone that a substance fluoresces is an important characteristic. The particular radiation that excites its fluorescence, along with the specific position of that fluorescence in the visible spectrum, can be identifying features. Additionally, fluorescence can help visually differentiate elements of a material that may otherwise appear similar.

Phosphorescence

Although fluorescence ceases almost immediately after the exciting radiation is removed, some substances will continue to emit luminescence for some time, even hours, after removal of the exciting stimulus. This phenomenon is called phosphorescence and is produced in compounds called phosphors.

Phosphorescence, like fluorescence, can be stimulated by many types of radiation. The image on an old television screen or an old computer color monitor (cathode-ray tube or CRT), for example, is produced by phosphorescence. An electron beam sweeps across the monitor screen and excites persistent luminescence in phosphors coated on the inside of the tube. Three different phosphors are present on the screen of a color monitor, and they emit red, green, and blue light to form colored images. Phosphorescence is still in use by analogue watch makers, in the form of a luminous dial that will “glow in the dark” to tell time. It is, however, a less common phenomenon and is not as generally relevant to cultural heritage imaging.

What is UV-induced Visible Luminescence (UVL)?

While visible luminescence induced from ultraviolet radiation can be observed with the human eye, UVL imaging is able to record the visible luminescence that is observed. UVL is used to differentiate materials (e.g., varnishes pigments in paintings, inks, glass formulations), enhance faded or obscured details, establish the condition (e.g., reveal mold and tidelines in paper objects), and reveal past conservation treatments (e.g., luminescing adhesives from repairs) (Mairinger 2000, Warda et al 2017, Webb 2019).

There are some available resources that provide references for the luminescence of materials related to conservation:

• Measday, Walker, and Pemberton (2017) compiled an extensive table available online with materials relevant to conservation and the luminescent characteristics.
• Mairinger (2000, p. 66) included a table of pure pigments powders (organic and inorganic) and their corresponding luminescing colors noting that paintings include layering and complex mixtures of pigments and binding materials that will impact the resulting luminescence and interpretation.
• The Kodak book Ultraviolet and Fluorescence Photography (1972) provided a table of pigments and corresponding luminescing colors that is included in the paintings conservation section.
• Grant (2000) provided examples of luminescence characteristics of art and artefacts for paintings, ceramics and glass, ivory and bone, metals, paper and parchment, textiles, wood, and minerals.

Several factors will impact the color and intensity of the luminescence response – the nature and amount of material, aging and degradation of the material, and the emission of the UV radiation source (Warda et al. 2017, p. 148). It is very important to note this when interpreting UVL results and consulting references reporting on the characteristic luminescence colors. Additionally, there can also be some confusion when both reflected light and luminescence are perceived simultaneously (Tragni 2005), which can result from parasitic and ambient stray radiation (Verri 2019). This underlines the importance of conducting the UVL imaging in dark room (to eliminate ambient stray radiation) and filtering the radiation source (to eliminate parasitic radiation).

Equipment

The three main components of an imaging system for ultraviolet imaging (both reflectance and luminescence) are (1) a recording device, or camera, (2) radiation sources, and (3) filters that can be used on both the camera and/or the radiation sources.

Cameras

UVL records images within the visible region and requires a standard digital camera.

Radiation Sources for RUV

See the above section on radiation sources. Note that sunlight cannot be used as a radiation source for UVL.

Filters

For UV imaging (both reflectance and luminescence), there are two different types of filters used: exciter filters and barrier filters (Mairinger 2000, p. 60). The exciter filters transmit long wave ultraviolet radiation and block visible light and IR radiation. Exciter filters can be used either on UV radiation sources for UVL or on the camera lens for RUV. Barrier filters transmit visible light and block UV and IR radiation in order to record the visible luminescence for UVL imaging. Filters are manufactured by different companies (e.g., Kodak, Schott, Peca, Hoya, B+W, etc.). While there are equivalent filters between different companies, these may vary in terms of transmission and comparing images acquired with different filters and setups may not be consistent (Chen and Smith 2020).    

Exciter Filter (Ultraviolet Transmitted Filter 365 nm peak)

An exciter filter is placed in front of the light source to allow only the desired exciting radiation to pass through while blocking other wavelengths. For example, in UV luminescence imaging, the exciter filter should transmit most of the UV light. In luminescence imaging with blue or blue-green light, the filter should transmit those wavelengths while blocking others, as stray light can overpower the luminescence signal.

Some UV light sources already have built-in exciter filters, though they may also transmit visible blue light. For sources without built-in filters, external filters like the Kodak Wratten Ultraviolet Filter No. 18A or Corning Glass No. 5840 can be used to screen out all but longwave UV and some near-infrared radiation. Modern UV-LEDs, like the realUV™ LED Flood Light, often include filtration at the diode level, eliminating the need for an external filter.

In some cases, shortwave UV radiation is needed for luminescence, such as when identifying certain materials or using chromatography. Filters like the Corning Glass No. 9863 can transmit shortwave UV down to 254 nm. For infrared luminescence imaging, an exciter filter must transmit visible light while blocking infrared, with filters like the Corning Glass No. 9780 being ideal.

Choosing the right exciter filter is crucial for successful imaging, and it should work in conjunction with a barrier filter to optimize image capture.

Barrier Filter

The exciter filter, placed in front of the light source, allows the necessary radiation to excite luminescence. However, not all of this radiation is absorbed, and some residual radiation can remain in the environment. If this residual radiation reaches the camera sensor, it will be recorded along with the luminescence, often overpowering the fainter luminescence signal. To prevent this, a second filter, called a barrier filter, is placed in front of the camera lens to block the residual exciting radiation.

An effective barrier filter should block all of the exciting radiation while transmitting only the light emitted as luminescence. For UV-induced luminescence, the barrier filter must absorb UV light. If the exciter filter transmits both UV and some short visible blue light, the barrier filter should block both UV and blue light.

The choice of barrier filter depends on the specific luminescence wavelengths. For example, if blue luminescence is present, the barrier filter must transmit blue light. However, if the blue luminescence wavelength is close to the blue light transmitted by the exciter filter, the exciter filter must be changed to one that blocks blue light. If the luminescence occurs at a longer blue wavelength, a filter like the Kodak Wratten No. 2E, which absorbs wavelengths below 425 nm, can be used. Additional filters, such as a visible bandpass filter (like the Peca 916 or 918), can help further refine the image. When working with a monochromatic sensor, any yellow filter that blocks UV while transmitting visible luminescence wavelengths can be used.

The spectrophotometric curves in the graph below show the transmittance and absorption of some common commercial filters.

The following is a table from Poseilov (2015, p. 106) and includes comparable filters that transmit UV (left column) and absorb UV (right column) for UVL.  

Exciter Filter / UV Transmitting Filters (on light source if necessary)		Barrier Filter / UV Absorbing Filters (on camera)
Useful Transmission range [nm]	Filter Designation	Maximum Cut begins [nm]	Filter Designation
320 - 385	B+W 403	390	Kodak Wratten 1A
270 - 375	MaxMax X-Nite BPI	400	Chance OW 12
330 (85% Peak)	MaxMax X-Nite 330C	450	Schott CG 8
300 - 400 (365 peak)	Kodak 18A		Kodak Wratten #2E
	Peca 900	500	Kodak Wratten #2B
	Schott UG1		Peca 702
	MidOpt BP365	520	Kodak Wratten #8
	Kopp 1041		Peca 708
250 - 480 (254 peak)	Kodak 18B	540	Kodak Wratten #12
	Chance OX 7		Peca 712
	Kopp 9863
	Corning 9863
250 - 380	Schott UG 5

UVL Standards, Targets, and References

Repeatability and comparability are challenging for multiband imaging. This is particularly challenging for UVL because of the different equipment being used, especially the range of radiation sources being used. Discussions around repeatability and comparability are not new and can be found in references like Dyer et al. (2013) and Webb (2019). The use of targets is one piece of improving repeatability and comparability for UVL.

Although the use of color and grayscale reference targets is common for visible light imaging, with several commercial options on the market and an array of software and workflows to choose from, this is not true for imaging in other areas of the spectrum. Very few commercial options are available and, of those that exist, many are quite expensive and the cost can be prohibitive for many labs and studios. Cultural heritage professional photographers, conservators, and scientists, often adapt targets or create custom references for their own purposes. At times, they must even resort to making bespoke targets to yield the desired results. These bespoke targets are then often used in combination with one or more of the above mentioned commercially available reference targets.

The commercially available targets include the X-Rite ColorChecker (and the AIC PhD could also be used), the UVInnovations Target-UV, the CHSOS TP-MSI calibration card, and spectral diffuse reflectance standards.

The X-Rite Passport ColorChecker is designed to profile a camera’s response to color under visible light conditions. It is not intended to be used beyond visible light and there are limitations with using it beyond visible light. However, it does contain a few patches that luminesce under UV radiation and this can provide a basic indication of exposure levels for UVL.

The UVInnovations Target-UV is a purpose-built UVL imaging target developed by conservators for conservation documentation (https://www.uvinnovations.com/target-uv). The target and the development of the target were presented at the 2014 AIC Annual Meeting, and a copy of the presentation is available on the UVInnovations website (McGlinchey Sexton et al. 2014). The UVInnovations website provides resources for workflows for using the target including determining the exposure and white balance. This target is intended for determining exposure based on the intensity of luminescence generated by different materials, but it was not designed to evaluate the overall imaging environment. Therefore it is usually necessary to also include a Spectralon or other similar target in the image.

[Add image of UV Innovations target]

Spectralon diffuse reflectance standards are also used for both UV luminescence and reflectance imaging. These standards are durable and chemically inert and provide a flat spectral response from UV into the IR (250-2500 nm). These are used for reflectance calibration for techniques like multispectral and hyperspectral imaging and they can be useful references for multiband imaging and techniques like UVL (as outlined in Dyer et al. 2013). The 99% diffuse reflectance standard provides information about the environmental lighting conditions for UVL (e.g., providing indication of stray or parasitic radiation).

[Add image of Spectralon]

The Cultural Heritage Science Open Source Technical Photography and Multispectral Imaging (CHSOS TP-MSI) Calibration Card comes with the purchase of CHSOS Pigment Checker and can be used for exposure and white balance for UVL images along with other multiband imaging techniques. The target includes shellac, madder lake, zinc white, and cadmium which have characteristic luminescence and can be useful for UVL imaging while the greyscale patches can be used for RUV. The CHSOS website includes exposure recommendations for the different multiband imaging techniques  (https://chsopensource.org/technical-photography-msi-calibration-card/).

[Add image of CHSOS TP-MSI Calibration Card]

It should be noted that many of these commercial options are fairly expensive, which is met by the development of bespoke/custom targets and simple, low-cost solutions.

For example, Pozeilov (2023) presented a simple and more economical option for UVL that can help to assess aspects of the imaging environment (i.e., camera filtration, radiation source emission, and ambient radiation) using PTFE tape and highly reflective metal sheet.

UVL Capture Considerations

In order to photograph an object using UV-induced visible luminescence imaging, the following steps are necessary and best done in a completely dark environment:

1. Use an ultraviolet radiation source to irradiate the object as evenly and uniformly as possible across its surface. An exciter filter may be required on the radiation source transmitting only the radiation needed to excite luminescence and blocking out all other undesired wavelengths.
2. Only visible light should be recorded by the imaging sensor inside the camera. Filter(s) are necessary on the camera lens to absorb radiation from the radiation source and transmit only the resulting luminescence from the object. Filters used on the camera lens are known as barrier filters.
3. The recording device can be modified or unmodified since the wavelengths that need to be recorded are within the visible range. Whether or not the camera is modified or not, filter(s) are still needed to ensure the emitted luminescence is reliably recorded by the camera’s imaging sensor.  

Different wavelengths of excitation radiation may all produce some degree of luminescence, but the intensity may vary significantly. Since changes in any of the luminescent intensities will alter the overall color, it is advisable to use the same radiation source throughout a study.  Certain details should be recorded during examination, including the type of ultraviolet source, the distance between the source and the object, and the distance between the object and the camera. For consistency, these distances should be kept constant. Since irradiation and other chemical alteration can influence luminescence, it is important to know the past history of the object as well.

UVL Exposure Determination and White Balance

While there are more established guidelines and best practices for exposure around reflectance imaging techniques, there is less available in terms of guidelines and best practice for luminescence techniques like UVL. Targets and standards can aid in exposure determination and white balance  in achieving consistent results.

The UV Innovations website includes workflow resources that provide guidelines for exposure and white balance when using the Target-UV (Capture Workflow 2018). The target does include references for different luminescence intensities (low, medium, high, and ultra) to cover a range of materials. It is noted that there are still some challenges with low intensity luminescence (McGlinchey Sexton et al. 2014), and also producing images that correspond to what is observed with visual examination (UV Innovations FAQ 2017; Webb 2019, p. 54).

The CHSOS TP and MSI Calibration Card comes with the purchase of CHSOS Pigment Checker and can be used for exposure and white balance for UVL images along with other multiband imaging techniques. The target includes shellac, madder lake, zinc white, and cadmium which have characteristic luminescence and can be useful for UVL imaging. The exposure recommendation is a value of 150 in the blue channel of the third grayscale patch from the left (https://chsopensource.org/technical-photography-msi-calibration-card/).

Case Studies

Ultraviolet Radiation Use in Paintings Conservation

Very often, before any attempt is made to clean, repair, or restore a painting, multiple imaging techniques are carried out to obtain all possible information about its condition. Ultraviolet imaging, both reflected and induced visible luminescence, are most useful in the examination of the painted surface and for evidence of recent retouches. It is also useful for locating residues of old varnishes and retouchings during cleaning procedures. Ultraviolet induced visible luminescence imaging provides information about any past retouching, overpainting, and restorations. After a painting has been cleaned and the old varnish completely removed, a luminescence photograph is often taken before revarnishing. This photograph will show the surface condition of the painting itself without the varnish barrier. Since varnish will absorb ultraviolet radiation and may produce a typical yellow-green luminescence, its presence on the painting would obscure any luminescence of the pigments.

When it is possible to photograph the actual pigment luminescence without its being masked by varnish, the color of the luminescence often aids in identifying the pigments. The following table contains a list of some pigments and their typical luminescence colors.

The placement of lamps for luminescence photography of paintings is the same as for any other photography of paintings. The diagram above shows a typical setup. In all luminescence photography, it is necessary to keep all other illumination from reaching the subject. Focus of the image can be accomplished with room lights turned on. They should then be turned off, however, for luminescence exposure. Windows must be covered with efficient shades or other means to exclude daylight. Exposure times for luminescence photography may range from several seconds up to an hour, depending on the intensity of the radiation source, the distance of the source from the painting, and the resultant luminescence brightness.