This page contains notes about photography of ultraviolet-A-radiation induced visible fluorescence. I discuss methods in detail and provide examples of higher plants, lichens and mosses. I describe the ultraviolet-A sources, UV-pass and UV-blocking filters, lenses and cameras that I use, as well as camera settings I use.
LED light, ultraviolet
Ultraviolet-A (UV-A) radiation at high irradiance can be damaging to eyes and skin, especially when there is little or no visible light together with it. Eye protection is recommended in all cases. It is safer to use only the so called “black light” corresponding to UV-A1 at wavelengths near 365 nm. However, even in this case when using flashlights or any other sources of UV-A that emits a concentrated beam, use of safety goggles is a must. When selecting goggles or safety spectacles pay attention to their UV-protection rating (e.g., UV400 CE, or increasing protection from ANSI U2 to U6 markings) as some goggles are certified to protect only from mechanical impacts (e.g., ANSI Z87+ with no U marking) (Figure 1). It is important that goggles and other safety eye wear provide protection also from the sides.
Clicking on photographs opens a gallery view with them displayed at a larger size and at higher resolution. This view also allows navigation among multiple photographs from a given Figure.
In this page code chunks can be “folded” so as to decrease the clutter. Above each plot, table or other R-code output you may find one or more “folded” code chunks signalled by a small triangle followed by “Code”. Clicking on the triangle “unfolds” folded chunks making the R code that produced the printed values, plots or tables visible. Clicking on the same icon on an “unfolded” chunk, folds it hiding the code.
The code when visible in the chunks can be copied by clicking on the top right corner, where an icon appears when the mouse cursor hovers over the code listing.
The </> Code drop down menu to the right of the page title makes it possible to unfold and fold all code chunks in the page and to view the Quarto source of the whole web page.
library(photobiology)
library(photobiologyFilters)
library(photobiologyLamps)
library(photobiologyLEDs)
library(photobiologyWavebands)
library(photobiologyPlants)
library(ggspectra)
library(patchwork)
library(knitr)
my.fig.heigth <- 4.5
my.fig.width <- 7
knitr::opts_chunk$set(
fig.width = my.fig.width, fig.height = my.fig.heigth, dev = "svg", out.width = "975%", message=FALSE
)
theme_set(theme_bw(12) + theme(legend.position = "top"))
photon_as_default()Fluorescence occurs when a molecule absorbs radiation and re-emits the absorbed energy also as radiation. The usual case is the capture of a single photon, dissipation of a (small) fraction of the energy as heat and the emission of the rest of the energy as a new photon. As the emitted photon carries less energy than the absorbed one, fluorescence takes place at a longer wavelength than the absorbed radiation.
Both UV and visible light can induce the fluorescence of chlorophyll in plants. Strong fluorescence is emitted at wavelengths near 685 nm and 740 nm, by chlorophyll-a and chlorophyll_b_. Phenolic acids fluoresce in the blue and green regions. Pigments in lichens, some mushrooms and even insects (as well as scorpions) fluoresce in a variety of bright pure colours.
We do not see the fluorescence in normal daylight as the high sensitivity of human vision with a peak in green prevent it. With a camera modified to increase sensitivity to infra-red radiation and a filter that blocks wavelengths shorter than 680–700 nm the fluorescence can be imaged. A flash of strong light is most effective and is used in the study of plant’s photosynthesis (Figure 2).
The fluorescence spectrum can be measured, with some difficulty with an array spectrometer and a UV-A flashlight resulting in very noisy data.
load("data/UVA365nm_source_003.spct.Rda")
autoplot(clean(smooth_spct(UVA365nm_source_003.spct, method = "supsmu", na.rm = TRUE)),
geom = "spct",
span = 55,
range = c(400, 800),
label.qty = "contribution.pc")
Of course it can be also measured with more sophisticated equipment to obtain a clean fluorescence spectrum, in this case for a wheat leaf, measured under 355 nm excitation (Figure 4).
autoplot(leaf_fluorescence.mspct, span = 11, geom = "spct")
The quantum yield of fluorescence depends on the molecules involved. In most cases the fraction of the incident radiation that is emitted as fluorescence is small, frequently not even detectable. When we illuminate a plant or object with ultraviolet-A (UV-A) radiation, only some of its components, if any, fluoresce strongly, emitting visible light (VIS) and NIR radiation of specific wavelengths.
To induce fluorescence across the whole visible spectrum we have to use radiation of shorter wavelengths, such as UV-A radiation. Our eyes’ low sensitivity to UV-A radiation and easily available eyeglasses and goggles that block all UV radiation help in allowing us to see the fluorescence without it being overwhelmed by the radiation used as excitation.
Different lichens contain pigments that fluoresce brightly when exposed to UV-A radiation (Figure 5). Colours of the fluorescence vary between yellow, red, occasionally blue and rarely green.
It can in exceptional cases when the energy acquired by a molecule by absorbing more than one photon is emitted as a single photon. In such a case, even if there is thermal energy dissipation, and the emitted photon carries less energy than the sum of the absorbed photons, it will carry more energy than each absorbed photon individually. This happens only in exceptional cases because the re-emission is normally fast.
I have in Flicker some galleries of fluorescence images.
Lichens photographed at the Jardin botanique du col du Lautaret, French Alps, 2019
Various photographs taken indoors, at home, Helsinki, Finland
Various photographs taken outdoors at night, Loppi, Finland, 2019-2020
In many respects photographing fluorescence is similar to photographing in low-light. It usually requires long exposure-times and use of a tripod or other camera support. In addition we need to make sure that what we photograph is really fluorescence emitted by the object being photographed rather than contamination by excitation radiation, ambient light or even fluorescence by nearby objects.
We can check if the light in the image is from the light source or not by including a “white” reflectance reference ([#fig-lichens-fluo-02]). I use a slab of white PTFE (Teflon) about 5 mm thick. It should have a matte surface and be clean, as dirt can easily fluoresce. (I use wet/dry sandpaper 600 grit under running water from time to time to clean it.) If the white slab appears darker, nearly black, in an otherwise brighter image, we can conclude that what has been recorded is fluorescence.
Some equipment and suitable surroundings are needed when photographing fluorescence. The items below are discussed in more detail in later sections.
Location, outdoors: At night, far from any artificial illumination in a moon-less night.
Location, indoors: In a dark room, with no strongly fluorescent objects except the one being photographed.
A source of UV-A radiation: A flash light based on a good UV-A LED is very convenient, especially outdoors. Indoors, mains powered LED or fluorescent “black light” lamps could be also used.
Exciter filter: A high-quality VIS-blocking filter is needed in the light source.
Barrier filter: A high-quality UV-blocking filter is needed on the camera lens.
Camera: A digital mirrorless (ML) system camera is preferable, but in principle any camera whose controls allow long exposures is enough. A converted camera is needed only to photograph NIR fluorescence.
Objective: Most lenses work, but a lens with an f/2.8 or wider maximum aperture is best.
Support: Exposure times between 20 s and a few minutes are common, so the camera must be mounted so that it does not move.
Subject: Easier to photographs are several lichens that fluoresce brightly, tolerate handling and do not wilt quickly. Of course, chlorophyll fluorescence can be strong under some conditions. Blue fluorescence is common in grasses.
As fluorescence is weak compared to other light sources it can only be observed and photographed in the absence of other sources of visible light. Outdoors, normally at night and far from street lamps and other light sources, or in a darkened room indoors. The main problem indoors is the presence of fluorescent objects other than those of interest. These include paper, light-coloured and white cloths and clothes, and even some plastic objects. The reason why UV-induced blue fluorescence is pervasive indoors is that blue-fluorescing compounds are normally added to white paper and some white plastics so that they look whiter and brighter. In fact, these chemical compounds are frequently called whiteners. Many laundry powders and liquids contain whitening and/or brightening agents that remain on clothes after washing them. Less frequently, white and light-coloured paints can be formulated with fluorescent substances in addition to white pigments that reflect visible light.
Post-it and similar yellow notes do not fluoresce much when exposed to UV-A radiation and can be used for labels or for taking notes.
I find UV-A LEDs most useful as UV-A sources as they intrinsically emit little VIS radiation. Little is anyway in most cases too much when photographing fluorescence and they need to be combined with optical filters that effectively block visible radiation. Flash lights and other small sources of UV-A are readily available as they are used for checking authenticity of bank notes, cleanliness (urine fluoresces strongly) and searching for scorpions in houses. LEDs emitting at a wavelength of 365 nm are the most suitable (Figure 7), as radiation at this wavelength is weakly visible.
LEDs emitting at longer wavelengths are more difficult to use because we see these wavelengths better, which makes seen the fluorescence more difficult. A key difficulty is finding a good enough barrier filter at say 450 nm or 480 nm. Such filters exist, but being special can be very expensive, but required to block violet and blue light that such LEDs emit.
LEDs emitting at wavelengths shorter than 360 nm tend to have much weaker UV emission, making them cumbersome for regular use. These LEDs are also more expensive and require more care to ensure safety.
autoplot(leds.mspct$LedEngin_LZ1_10UV00_365nm, geom = "spct")
Although the emission of LEDs takes place in a relatively narrow peak, the tails can extend tens of nanometres on either side of the wavelength at the maximum emission (Figure 8). Furthermore, the peak wavelength can shift by a few nanometres as the LED temperature and/or current used to drive it changes.
autoplot(leds.mspct$LedEngin_LZ1_10UV00_365nm, geom = "spct", range = c(315, 450))
A flashlight with a Nichia UV-A1 LED with nominal peak of emission at 365 nm, filtered with a VIS blocking filter has a much shorter right-side tail.
autoplot(smooth_spct(normalise(lamps.mspct[c("Convoy.S2plus.LED.UVA.flashlight")], norm = "undo"),
method = "supsmu", strength = 0.001), geom = "spct",
range = c(315, 450))
Black-light-blue and UV-A fluorescent tubes can also be used indoors, or where mains power is available. They usually emit over a broader range of wavelengths than individual UV LEDs, something that can sometimes be an advantage (Figure 10). Many of them emit a significant fraction of the total radiation in the visible. This makes fluorescent tubes much less suitable as blocking the unwanted VIS and NIR radiation is more difficult than with UV-A LEDs.
autoplot(lamps.mspct[c("QPanel.FT.UVB340.40W", "Eiko.F36T8.BLB")],
geom = "spct", idfactor = "Lamp", facets = 1, span = 101)
In principle a photography flash modified to emit UV-A radiation could be used as a light source. Xenon lamps do emit over a very broad range of wavelengths, making them very difficult to filter effectively (Figure 11).
autoplot(lamps.mspct$Godox.XeF.AD200.H200.flash, geom = "spct", span = 51)
To be able to photograph fluorescence isolated from VIS and NIR radiation, without contamination from the UV-A sources, it is necessary to use UV-pass VIS-blocking filters on the light source. This barrier filter should completely block the excitation light making a high-quality UV-blocking filter necessary. Most UV-filters sold for photography, even from well-known brands, have a cut-in at too short wavelengths (330 to 380 nm). However, when using a filtered 365 nm LED as light source, the filter used on the lens has to block wavelengths up to just past 400 nm.
The best, but rather expensive, option is a Zeiss UV T* filter, second best is a significantly cheaper Firecrest UV400 filter (from Formatt-Hitech, apparently gone bankrupt), and a third option is a Tiffen Haze 2A filter (rather difficult to find outside USA). The first two are interference filters and reflect UV, rather than absorb it. The Tiffen filter, is of a rather old type and in the tradition of Kodak Wratten filters, has a thin UV-absorbing gelatine layer encased between two glass sheets. None of these three filters fluoresce significantly when exposed to UV radiation. In contrast, as discussed later in this post, most yellow, orange and red filters are made of ionic glass and when illuminated with UV radiation fluoresce strongly. Thus, they can be used only behind a UV-blocking filter. Common NIR pass filters used in photography are also ionic and do also fluoresce, but less intensively.
autoplot(filters.mspct[c("Zeiss_UV_Tstar_2.0mm_52mm",
"Hoya_UV0_HMC_2.0mm_52mm")],
range = c(315, 450)) + theme(legend.position = "top") 
When combined with the emission spectrum of the Convoy 2+ flashlight, the Zeiss filter “leaks” only \(\approx 0.03\%\) of the UV from the flash light, while the Hoya filters “leaks” \(\approx 6.5\%\) (Figure 13). This leak is more than enough to interfere with the detection of fluorescence. How much of this radiation reaches the camera sensor depends on the lens, but in any case it would be a major problem.
(autoplot(smooth_spct(normalise(lamps.mspct[["Convoy.S2plus.LED.UVA.flashlight"]], norm = "undo") *
filters.mspct[["Zeiss_UV_Tstar_2.0mm_52mm"]], method = "supsmu", strength = 0.5),
range = c(315, 450), ylim = c(NA, 6.2),
annotations = c("-", "peaks")) + ggtitle("Zeiss UV T* filter") +
theme(legend.position = "top")) |
(autoplot(smooth_spct(normalise(lamps.mspct[["Convoy.S2plus.LED.UVA.flashlight"]], norm = "undo") *
filters.mspct[["Hoya_UV0_HMC_2.0mm_52mm"]], method = "supsmu", strength = 0.5),
range = c(315, 450), ylim = c(NA, 6.2),
annotations = c("-", "peaks")) + ggtitle("Hoya UV(0) filter") +
theme(legend.position = "top")) + plot_layout(axis_titles = "collect")
UV-A flashlights are convenient excitation sources for use in the field (Figure 15, Figure 16). When the working distance is relatively short or even illumination is desired, a flood flashlight works better. In the case of long exposure times and use of “light painting” the more concentrated beam from the Convoy 2+ allows better targeting of the UV radiation. The Convoy 2+ is also suitable for illuminating small objects. Both flashlights have Nichia LEDs with a nominal peak of emission at 365 nm. The exact types were not specified when I bought the flashlights, but at least the Jaxman Uc1 most likely has a NVSU233B(T) LED from Nichia.
The emission spectrum of Convoy 2+ flashlight ([Figure 15) and of the Jaxman Uc1 flashlight (Figure 16) have no stray light at longer wavelengths. Both flashlights use a single Lithium battery of type 16850, that needs to be removed for recharging.
autoplot(lamps.mspct$Convoy.S2plus.LED.UVA.flashlight, geom = "spct")
autoplot(lamps.mspct$Jaxman.U1c.LED.UVA.flood.flashlight, geom = "spct")
There are many different UV-A or black-light flashlights available. Most suitable are those emitting at 365 nm, because those emitting at 385 nm and longer wavelengths emit much more visible light (Figure 17), that will either interfere with the recording of the visible fluorescence or be blocked by the necessary UV+blue-blocking filter. The flashlight in this example emitted much less radiation and at longer wavelength.
It is also important to consider the beam uniformity and breadth. Finally, the power of most flashlights is not given by sellers as energy flux of UV-A radiation but instead as electrical power consumed. In some cases the power values given are clearly inflated, possibly based on the maximum power the LEDs withstand rather than the power at which they work in the flashlight or are simply imaginary, i.e., a hand-held flash light with a rating of 200 W if true would burn the user’s hand in no time!
autoplot(lamps.mspct$Generic.LED.UVA.flashlight, geom = "spct")
The Convoy 2 and 2+ flashlight have been available from some years. They are modular, available in many different configurations and even as parts for self assembly. Flashlights using the same metal body and other parts are sold under other brands. For example, the Jaxman is very similar to my Convoy 2+ but it has, apparently, a different LED and or optics, and a different LED driver.
The same body Convoy 2+ is sold with different LEDs installed, different wavelengths, different powers, and with different packages. There are two types of reflectors, clear glass, AR-coated glass, and UV-pass VIS-block windows. Drivers differ in the maximum current, that needs to be matched to the LED used, and in the user interface. The user interface can allow dimming or not, and even in some cases have a blinking option. Some drivers have a single user interface and others have multiple ones to choose from. One has to be careful with the driver and LED, as a driver that feeds a very high current to the LED will heat the flashlight. Several of the drivers have over-heating protection that decreases the power as needed to avoid overheating. This works when use is briefly, but for a longer photography session one want constant illumination.
All in all it is best to buy a ready assembled Convoy flashlight with Nichia UV LED with peak at 365 nm and an UV-pass filter. It is possible, however, to buy the parts separately, even from different suppliers and assemble a custom version with a specific LED of one’s choice. When I bought the Convoy 2+ it was not available with a pre-installed filter. Nowadays there are multiple UV-A versions, and those using LEDs from other brands than Nichia are cheaper.
Configurations ready available are with white light of different colour temperatures and powers, UV-C. UV-A, blue, green, read and infra-red. Convoy has an official store in AliExpress (“Convoy Flashlight Store”) that has this and several other flashlights.
The standard configuration uses a 18650 rechargeable Lithium battery. The body of the flashlight has a diameter of about 20 mm and is 80 mm long. To hold it I am using a holder that is sold to attach small cameras to the handlebar of a bicycle. It can be found in AliExpress under the name of “1/4 Screw Metal Cycling Bike Mount Motorcycle Handlebar Holder for GoPro 13 12 11 10 9 8 Insta360 X3 X4 DJI Osmo Action 4 5 Pro” by multiple sellers. This adds as attachment point a 1/4”-20 male thread, the standard size used for cameras. In the photograph, I have attached a male “flash cold shoe”.
In the second half of 2025 a UV-A1 flashlight model with zoom optics has become available both through AliExpress sellers, very cheaply at \(\approx 8\) €, and through a supplier of equipment for UV and IR photography in the USA at a higher price of \(\approx 20\) €. A zoom optics has a rather thick lens instead of a thin glass window or filter. To be useful in UV-A1 a lens needs to transmit wavelengths down to 340 nm well. I haven’t found any spectra or specifications for the LED or its power. The cheap and expensive flashlights look identical in the seller’s photographs. Flashlights with zoom optics have earlier been available for visible light and NIR radiation. I have one of each, and specially the white light one from LedLenser is excellent and very convenient to use.
The brand of the cheap ones is AloneFire, and reviews of another flashlight of this brand are negative, indicating badly designed electronics and cooling. The higher priced version from Kolari Vision could potentially have different electronics in the same body. By comparison, the Convoy 2+ with 3W Nichia LED costs \(\approx 35-45\) € in AliExpress and the Jaxman Uc1 with 6W Nichia LED and flood optics \(\approx 70-75\) € in AliExpress.
As mentioned above, if we want to photograph fluorescence across the whole visible range, the light source used for excitation should not emit any radiation in the visible range. In fact, more generally we need to use complementary filters in the light source and camera. Most frequently we aim at a cross-over of the transmittances so that neither of the two filters transmits at 400 nm. In practice, we need a “dead-zone” as the cut-off and cut-in wavelengths of filters have tolerances of a few nanometres and there can be a gradual change in transmittance within a range of wavelengths. We look first at the filters commonly used on light sources or “exciter filters”.
Digital cameras have built-in filters on their sensors, but these filters frequently let through some UV-A radiation. Additional filters must be chosen considering the light source in use. For example, many UV-pass filters are not effective at blocking the near infra-red radiation (NIR) that some cameras can “see” (Figure 18). Fluorescent lamps emit some NIR radiation while UV-A LEDs do not emit it (cf. Figure 10, Figure 7).
autoplot(filters.mspct[c("Tangsinuo_ZWB1_2.1mm_52mm", "Tangsinuo_ZWB2_2.0mm_52mm")],
geom = "spct", idfactor = "Filter", span = 51)
Filter types ZWB1 and ZWB2 from various Chinese suppliers are cheaper than the German made Schott UG1 and UG5. Specifications are less tight for the Chinese filters, and can sometimes present optical defects. ZWB1 and ZWB2 are good filter types for use with UV-A LEDs. ZWB1 with a cut-off at a slightly shorter wavelength can be preferable. A filter thickness of 2 mm or more is necessary. These filters are usually fine if bought from reliable suppliers. Filters ready cut to size for common flashlight types are readily available through AliExpress and eBay sellers. In LED UV-A light sources it is best to use a VIS-blocking filter.
If we are interested in red or NIR fluorescence we can use blue- or UV + blue radiation for excitation, with suitable filters and LEDs or lamps.
The Convoy 2+ flashlight is available in innumerable variations, with different LEDs, different LED drivers, and different reflectors. Even the parts can be bought as a key, and as it uses a standard sized metal core printed circuit board for a 3W LED, it would be possible to assemble one with almost any SMD 3W LED, or even 6W LED. However, instead of assembling one from parts, I bought a ready-made one with a high power blue LED.
autoplot(lamps.mspct$Convoy.S2plus.LED.blue.flashlight, geom = "spct")
As with ionic absorptive UV short-pass filters, blue short-pass filters block NIR very poorly (Figure 20). So, once again, easiest is to use LEDs as blue-light excitation sources.
autoplot(filters.mspct[c("Tangsinuo_ZB1_2.0mm_52mm",
"Tangsinuo_ZB2_2.0mm_52mm")],
geom = "spct", idfactor = "Filter",
annotations = c("+", "wls"),
span = 101)
Differences in blocking efficiency among reflective dichroic filters can be huge (Figure 21). The differences in price are also huge, in the case of these two filters, 20 € compared to 190 € for filters 25 mm in diameter.
autoplot(filters.mspct[c("Thorlabs_FBH450_40", "UQG_Blue_dichroic_CDB")],
geom = "spct", idfactor = "Filter", range = c(300, 1050),
annotations = c("+", "wls"),
span = 101)
Barrier filters are used in front or behind the camera lens to block the excitation light completely. In this case there is an additional difficulty, yellow, orange and red glass filters themselves fluoresce when exposed to UV radiation. The solution is to use a UV-blocking filter that either reflects UV-radiation in front of the glass filters (Figure 22). These UV-blocking filters can be, as discussed above, dichroic interference filters or filters based on a different pigment than the metal ions used in solid glass filters. Kodak publications from the film era recommend using a Kodak Wratten 2A or 2B filter in front of other barrier filters. Tiffen still sells filters very similar to the Kodak ones. In these filter a coloured light-absorbing film, in most cases using gelatine as medium for the pigment, is encased between two layers of clear optical glass. These filters used to be also available as bare gelatine films. The Zeiss UV T* filter is based on interference and reflects radiation below 405 nm, it is AR coated and less affected by reflections at the transmitted wavelengths than the Tiffen. The Zeiss T* filter prevents the fluorescence of other glass filters as effectively or better than the Tiffen Haze 2A, but it is rather expensive. The Firecrest UV400 multi-coated filter is nearly as good as the Zeiss, but significantly cheaper. In actual use, these three filters work well.
autoplot(filters.mspct[c("Tiffen_Haze_2A_2.6mm_52mm",
"Zeiss_UV_Tstar_2.0mm_52mm")],
geom = "spct", idfactor = "Filter",
annotations = list(c("-", "peaks"), c("+", "wls")))
In some cases we may want to photograph fluorescence only of some colours. High quality, dichroic band-pass filters would be perfect for this, but at reasonably large diameters, are very expensive. These filters are the ones used to separate fluorescence from different dyes in microscopy. If the aim is to detect all florescence at wavelength longer than a target, it is possible to use a long-pass filter. If this is a ionic glass filter, mounted behind a UV-blocking filter (Figure 23).
autoplot(filters.mspct[c("Hoya_Y_(K2)_HMC_2.3mm_52mm",
"Heliopan_Yellow_5_SH_PMC_2.3mm_52mm",
"Heliopan_Orange_22_SH_PMC_2.2mm_52mm",
"Heliopan_Red_25_SH_PMC_2.2mm_30.5mm",
"Heliopan_RG665_2.3mm_46mm")],
geom = "line", idfactor = "Filter",
annotations = list(c("-", "peaks"), c("+", "wls")))
With a camera modified to “see” a broader spectrum by replacement of the sensor filter, it is possible to photograph NIR fluorescence by blocking visible fluorescence with long-pass barrier filters (Figure 26). Compared to an unmodified camera leaks of NIR from the excitation radiation source become an even bigger problem. Also NIR long-pass filters require “protection” from UV to prevent their fluorescence.
Yellow, orange, red and NIR long-pass glass filters fluoresce when exposed to UV-A radiation, most strongly the yellow filters (Figure 24). Yellow filters do emit a very strong yellow fluorescence. The fluorescence of orange and specially red filters is not so intense. The fluorescence of NIR filters is mainly NIR and not visible to a normal camera or the naked eye. Using a full-spectrum converted camera and a long-pass filter on the lens, we can see that all four glass filters fluoresce in the IR.
The spectra of the fluorescence emission have one or two peaks depending on the filter, and these peaks are at longer wavelengths than those of the radiation used for excitation (Figure 25).
names(filters_UVIVIF.mspct) <- gsub("_|52mm$", " ", names(filters_UVIVIF.mspct))
autoplot(filters_UVIVIF.mspct,
facets = 2,
w.band = NULL,
annotations = c("-", "labels"),
span = 21)
To control fluorescence from these filters an additional non-fluorescent UV-reflecting or absorbing filter must be stacked in front of them. Examples of effective UV-blocking filters are Zeiss UV*, Firecrest UV400 and Tiffen Haze 2A. Fluorescence can be also excited by blue light, so UV-blocking filters do not always prevent fluorescence from yellow, orange and red glass filters.
autoplot(filters.mspct[c("Heliopan_RG695_2.2mm_52mm",
"Heliopan_RG780_2.3mm_52mm",
"Hoya_R72_2.4mm_52mm",
"Zomei_IR850_2.1mm_52mm")],
geom = "line", idfactor = "Filter",
annotations = list(c("-", "peaks"), c("+", "wls")))
In principle band-pass filters can also be used, although this is a specialized situation. Of band-pass filters, the only sold for photography ready mounted on rings are the UV-IR cut filters. The actual wavelength boundaries vary rather broadly among brands and types (Figure 27). In AliExpress and eBay sometimes unusual ones are offered.
autoplot(filters.mspct[c("Firecrest_UVIR_Cut_0.96mm_52mm",
"Heliopan_UVIR_CUT_Digital_2.2mm_52mm",
"Rocolax_UVIR_Cut_445nm_650nm_1.1mm_52mm",
"Rocolax_UVIR_Cut_PRO_HD_(W)_1.1mm_52mm",
"Fotga_UVIR_CUT_0.54mm_52mm")],
geom = "line", idfactor = "Filter", facets = 1,
annotations = list(c("-", "peaks", "summaries"), c("+", "wls")))
Filters can be stacked, and this is frequently necessary with UV-pass filters that also transmit some NIR radiation.
A huge selection of band-pass filters mounted on suitable rings are available from suppliers of equipment for industrial machine vision. For example, MIDOPT (distributed in Europe by Stemmer Imaging), is a well known supplier, but such filters tend to be more expensive than those sold for photography. Another reliable supplier with a large variety of filters is Thorlabs, including filters specifically made for fluorescence imaging in microscopy, but mainly in smaller sizes. Square filters in a 2” \(\times\) 2” or 50 mm \(\times\) 50 mm size are common in the catalogues of optics suppliers like Hebo glass, UQG Optics, and some local optics shops, e.g., Teknofokus in Finland.
Normal cameras can be used for photographing the visible fluorescence. They are preferable to full-spectrum converted cameras as the built-in sensor filter help block unwanted UV and IR radiation. To photograph NIR fluorescence, an IR or full-spectrum-converted camera is needed.
A key requirement is that the camera supports long exposure times, i.e., that it has at least a Bulb setting for shutter speed as exposures may extend to minutes. Cameras that can display the live image as is “builds” during a long exposure are extremely convenient. In Olympus/OM-System cameras this mode is called and is what I always use for UVIVF photography in the field at night. In the case of close-up and macro photography it is possible to use exposure times that are shorter and to illuminate the whole view evenly.
A lens with a large aperture helps, but is not a must as long as subjects do not move. Any good lens is suitable as the UV radiation is used to excite the fluorescence and we block it with a filter before it enters the lens. Thus, it does not require the especial lenses and modified cameras that are required for reflected UV photography.
Objectively white balancing a photograph of fluorescence is nearly impossible. After trying other approaches I have mostly settled into editing the colour in these photographs to “look right”, i.e., matching my recollection of how the subject looked like when I photographed it. However, in some cases I use just a daylight balance and alternatively edit the colour to remove intense casts so as to increase the apparent colour range.
Combining the close-up or macro-photography with ultraviolet-induced visible fluorescence in the field can be difficult as exposure times are long and focus stacking is frequently needed for subjects with relief. A few examples from last night can be seen in Figure 28.
With suitable filters and lenses we can even attempt macro-photography of UV-A-induced chlorophyll fluorescence with a modified mirrorless camera (Figure 29).
Photographing UV-A-induced visible fluorescence is interesting and less demanding in equipment and technique than photography in reflected UV radiation. An unmodified digital camera can be used and long pass filters are in general cheaper and more common than UV short-pass filters that block visible and infrared radiation effectively enough. UV-A as excitation is convenient and easy to work with, blue light could be used instead if we are interested in fluorescence at longer wavelengths or for safety reasons in teaching situations. Using UV-B radiation to induce UV-A and violet fluorescence is in principle possible but more demanding in practice. Photographing visible-light induced near infrared fluorescence is more demanding in equipment, but opens interesting possibilities. Could we with comparatively cheap equipment use photography to qualitatively assess changes epidermal UV transmittance under controlled conditions?
The UltravioletPhotography discussion group is a very friendly and knowledgeable group of people from around the world with an interest in UV, NIR, and UV-induced fluorescence.
The web sites of Adrian Davies, author of three of the books listed below, include galleries of photographs, including some with photographs of UV-A induced fluorescence. See Imaging the Invisible and Adrian Davies imaging.