Dressed in uniform, wearing latex gloves and with a never smiling face, the security officer at the airport tells you to take off your shoes, empty your pockets and put everything containing metal or electronics into a tray and put it on the conveyor belt to be x-rayed while you step through the bloody body scanner. If you are like me, this means that you have to open up that camera bag and put it in a tray. And this is where the worries that keep you up at night start. Should you ask for a hand check? The security officer says that a single scan won’t do any harm for films up to EI 1600. You console yourself with the thought that they must have tested this… Right..? The doubt sets in again. Will your film come back home unharmed and without ill-effects from the x-rays? What if the pictures you took will be ruined when you get home!?

Earlier this year I went through the same horrors on a trip to Vietnam, on which I encountered multiple security checkpoints with x-ray scanners. Some of the devices were state of the art, while others seemed to be leftovers from Soviet manufacture. I was lucky, and encountered only helpful security personnel that took little persuasion to get them to check my films by hand. Nonetheless, I have encountered airports in other countries in the past where a hand check was refused for various reasons, which made me wonder how many scans your film can take before defects will become apparent.

In this article, I explore the theoretical and scientific side of x-ray exposures to photographic film, show the results of the field test that I performed with black and white film and conclude with some advice. Due to the extent of the tests and the rigor I want to include here, this article might be a tough cookie to get through, so bear with me. If you are only interested in the summary, click here now. If you are interested in all the nitty-gritty details, scroll down and keep on reading. I hope there is plenty to satisfy your curiosity.

Table of contents

  1. What do the film manufacturers recommend?
  2. How can x-rays harm my photos?
  3. The field-test: shoot film and x-ray it
    1. Plan A: the controlled test
    2. Plan B: no control and hope for the best
    3. Film stock
    4. Is 35 mm film shielded by its encasing?
    5. Test scene
    6. Development
    7. Visual inspection and densitometry
    8. Printing
  4. Conclusions
  5. Summary
  6. Acknowledgements

What do the film manufacturers recommend?

Table of contents

Before we go into the details, it is instructive to consider the recommendations from the film manufacturers themselves. They tested how their films fare under x-ray exposure, obviously. Film manufacturers know (or at least used to know) their film best and I found it hard to believe they would not have conducted any tests regarding the effects of x-ray exposure to professional photographic and cinematographic film. Unfortunately, with the exception of some documents giving well intended advice on traveling with film [5, 6], I could not locate any other documents outlining tests or test results published by the manufacturers themselves. Not on the internet, nor in the scientific literature. The only concrete discussion of research conducted by manufacturers that I came across was hidden in a forum thread that pointed at the I3A.

The International Imaging Industry Association (I3A) is/was a non-profit organization, that counts major film manufacturers among its members. According to user bmattock of the rangefinderforum [7], the I3A conducted tests that included passing film up to 100 times through x-ray scanners to determine the maximum number of passes that can be sustained without ill-effects. Unfortunately, the I3A website is out of use and the documents cannot be accessed directly anymore.

Enter, the Internet Archive.org’s WayBackMachine. Using this web-archive, I was able to recover two short reports on the tests that the I3A conducted on the InVision Technologies CTX-5000SP x-ray scanners [8, 9]. The documents report on tests of 1, 5, 10, 25, 50 and 100 consecutive inspections by the scanner in its low-power mode. Some of the tests included the use of the high-energy beams that are used for the detection of explosives. They chose this specific scanner model, because the American Federal Airworthiness Association (FAA) purchased and certified over 150 of these devices in the late 1990s and the early 2000s. This exact purchase prompted the tests by the I3A, because these devices contained x-ray components working at different intensities than those that came before them. Especially the high intensity beams were suspected to potentially harm film.

The tests were conducted by representatives of Eastman Kodak, Fuji Film, Ilford, Konica Imaging and the FAA. The reports, which can be downloaded from here (1) / (2) or via the WayBackMachine (1) / (2), conclude the following:

  • When only the low energy system was used (of unspecified intensity or wavelength), no perceivable fogging occurred even after 50 passes, although it could be picked up by sensitometers. As technicians of the involved firms could not perceive any ill-effects by visual inspection, “average consumers” were expected not to notice any effects. The report does not speak of effects that could be noted by professional users.
  • The high intensity beam leaves a distinct line on the film at all tested sensitivities as low as EI 100. The claim that x-ray machines would be safe up to EI 1600 thus highly depends on the configuration of the device, but should definitely not be taken for granted.
  • All color negative film will sustain significant fogging when hit by the high intensity beam of the CTX-5000SP. Whether you will notice any effects on the print may vary with image subject, as busy scenes may obscure its effects.
  • For color negative film a loss of contrast and an increase of granularity is noticed after 10 passes (and may thus occur between 5 to 10 passes). This is marked in red as a significant result, because it does affect print quality. In addition to this, minimal color shifts were recorded. This seems to contradict the first claim that mentioned no perceivable fogging up to 50 passes.
  • EI 400 black-and-white film shows banding already after 5 passes through the machine and the manufacturer – likely Ilford, as it reports the use of Ilfotec DD as the developer of choice – reported noticeable fogging at 5 to 10 passes for both EI 400 and EI 3200 film.
  • EI 3200 black-and-white film shows fogging at 1 – 5 passes and clear banding at 25+ passes.

They continue by recommending that a 5 pass maximum should be maintained for color negative and color reversal film up to EI 800, and that film of higher sensitivities and all black-and-white films should always be checked by hand.

How can x-rays harm my photos?

Table of contents

To answer this question, we need to understand how the latent image in the emulsion is formed. After some gross simplification, high school physics should be sufficient to understand this process on a conceptual level. You may have long since been forgotten high school physics, though, so please bear with me in the following sections.

We use the Gurney-Mott-Webb model, which is discussed in greater detail in another article. This model assumes that the emulsion consists of silver bromide grains and a small quantity of silver sulphide. Within the grains (shown in Fig. 1), the positively charged silver ions and negatively charged bromide ions are arranged in a lattice. By an external trigger, for example by the absorption of light, the silver bromide dissociates and an electron can become free to move within the grain. If it manages to reach a silver sulphide speck before being captured on its way, it is absorbed by the silver sulphide, which hereby becomes negatively charged. The negative charge causes an attractive force on the positively charged silver ions. Defects in the lattice allow sufficient freedom for the silver ions to move towards the silver sulphide speck. When one silver ion reaches its destination, it absorbs the excess electron and metallic silver is formed. When sufficient silver is formed, it becomes developable and will result in a visible darkening of the negative after development.

Figure
Figure 1: a grain in the emulsion contains a lattice of positively charged silver ions (gray) and negatively charged bromide (green) ions. Silver sulphide specks (orange) are present in the gelatin support and the emulsion. When light hits the grain, electrons are liberated and are absorbed by the silver sulphide speck (b). Defects in the lattice allows the silver ions, which are attracted by the negatively charged speck, to move and realize silver clusters to be formed near the speck (c). This figure was previously published here.

To understand how the absorption of light results in freed electrons, we need to zoom in to a single atom. Every atom consists of a nucleus which is surrounded by a cloud of one or more electrons. The distance at which the electrons orbit the nucleus is set by their energy: the higher the energy, the further removed they will be. It turns out that only certain values are allowed and that there are only discrete levels possible (or in other words, the energy levels are quantized). When an incoming photon has the right amount of energy, it is absorbed and its energy is used to raise an electron into a higher orbit. Visible light is energetic enough to raise electrons from their bound state, in which they stay with their associated ion, to the conduction band, in which they are free to move between ions in the lattice [10, 11]. It is these electrons that wander in the silver bromide grains and are the vital link in forming a latent image. Note here, that there is little to no amplification in this process: a single photon results in a single conduction electron, and thus one opportunity to form metallic silver.

Other types of radiation, such as alpha and beta particles, can also provide sufficient energy to set off this and similar processes. More important for the present discussion, however, are the types of electromagnetic radiation that cannot be seen by the naked eye: ultraviolet (UV), infrared (IR, also known as radiative heat) and x-rays [12]; the one more energetic than the other, depending on the wavelength. This energy is typically expressed in the common unit of energy eV, which stands for electronvolt.

Visible light has photon energies in the order of 1.59 eV to 3.26 eV; the x-rays used in luggage scanners have photon energies in the order of 350 keV (read: kilo-electronvolt). In other words: the used x-rays are over 100.000 times more energetic than visible light. When such a photon hits the silver bromide grains it is not just absorbed. It is so energetic that it instead knocks the electrons out of their orbits around the atom core and jettisons them into the material. This effect is known as the photoelectric effect.

Such a jettisoned electron, known as a photoelectron, can travel through many grains in the emulsion before it is absorbed again. On its way it encounters many silver bromide molecules, which would otherwise remain unaffected by the incoming radiation. The photoelectron loses its energy by ionizing these molecules, which in turn all release conduction electrons that can form a latent image in the same way as explained for the case of visible light. Because the photoelectron can cause tens to hundreds of thousands conduction electrons before having lost all its energy, it is very efficient in creating many of them. To those that have read the article on low intensity reciprocity failure, it should not be surprising that we do not see such effects for x-rays [13]. Note here, that there is an amplification in this process: a single x-ray photon results in a single photoelectron, which sets off a chain reaction and launches many conduction electrons that can form metallic silver.

Interestingly enough, per unit of energy x-rays are less efficient in producing density than visible light [12, 14]. When visible light strikes the grains, the conduction electrons can only travel short distances and result in the formation of silver close to where the photon was absorbed. The high energy photoelectrons formed through x-rays, however, travel across grain boundaries and disperse far in the emulsion. Although it causes the release of many conduction electrons, it does so over comparatively large distances. The formed silver is then also spread out over a large area, resulting in a low density per unit of energy. Moreover, the very large number of conduction electrons that are simultaneously working to form metallic silver can saturate the image forming process in a way that is similar to the saturation effects found in high intensity reciprocity failure.

After all this science, what does this mean for our undeveloped photos when they are inspected with an x-ray scanner? At low scanner intensities an inspection will result in a fogged image. The negative will gain additional density nearly uniformly over the entire frame. Uniform, slight fogging is not a problem in practice, because it can be ‘printed through’. In the darkroom it can be compensated for by lengthening the exposure time of the enlarger. When sufficient density has built up, however, this will start to affect image contrast as shadows are less affected by this than the bright highlights. When a high intensity x-ray beam hits the emulsion, locally high densities can be created which show up as narrow, washed out lines. They are typically curved, as the beam passes over the film while it is rolled around its spool. In busy scenes these may be obscured by the density that is already in the photograph, namely that makes up the actual image, but these lines do affect image quality directly.

The field-test: shoot film and x-ray it

Table of contents

Browsing the (scientific) literature and the world-wide web, only very few examples pop up that visually demonstrate the effects of x-ray exposure on the developed images. Most of the information you can find on the internet is anecdotal or refers back to the I3A reports. As the old saying goes: a picture says more than a thousand words. However, this requires a test under realistic conditions to get these images. Such a test requires two things: a large quantity of film and access to an x-ray luggage scanner.

Plan A: the controlled test

Table of contents

A large quantity of film is easily obtained, albeit nowadays at slightly high expense. Access to an x-ray luggage scanner proved a lot harder to come by. I have reached out to 7 regional airports across the country, a school that trains airport security personnel, a security company that specializes in airport security and a company that rents out x-ray scanners to security companies. Some of the larger airports kindly declined my request for a face-to-face meeting and do not allow me to conduct tests in their equipment, while others never replied. The reasons they gave me ranged from safety concerns and the busy holiday season coming up, to legislation of the Ministry of Security and Justice that prohibits testing or experimentation using x-ray machines. They kindly refused me again, even when I told them I do not need to see or be near the machine, as long as the films are scanned systematically. With the increased severity of security measures lately, I cannot blame any of them for not wanting to take any (perceived) potential risks.

A friend working at a company that manufactures x-ray inspection tools for the food industry offered to scan the film for me, but their tools use different energy levels and different dosages, which makes these devices not suitable for representative testing. It also means that I would have to mail exposed film to the UK, which then might be scanned by a cargo scanner before it gets on the plane. Cargo scanners are much more powerful than luggage scanners (3 MeV instead of 300 keV), and are considered to be a no-go from the onset [5].

Plan B: no control and hope for the best

Table of contents

Without direct access to the right kind of x-ray scanner, a test with controlled and known conditions is out of reach. Plan B, therefore, is a long-term project that involves many flights. That is not something I can realistically afford myself. Luckily, I have plenty of colleagues that fly regularly and that were willing to help me out by taking a few rolls of film in their hand luggage. By distributing the rolls over several people and several flights, all rolls could get the required number of x-ray inspections within a time span of roughly two months.

Although some of them are amateur photographers like me, none of them are familiar with using film. Therefore, I instructed them to handle the film with the same care as they would handle a bar of chocolate: keep it in your hand luggage and away from high temperatures or direct exposure to light. Each of them carried 2 to 4 rolls in a transparent plastic bag in their hand luggage. I chose to use these bags in case airport security would require an additional hand check and to protect the films from moisture and dirt. It also conveniently kept the films together, so that they wouldn’t get lost in someone’s bag.

This scheme has several consequences. Most importantly this means that the test conditions are unknown and uncontrolled. As a scientist, this is something that makes me shrivel a bit, but it is a direct consequence of the limited resources. Also, I have no knowledge of how the films were handled, and if they were scanned once, twice or for prolonged times in a single inspection and have to rely on the accounts of my colleagues. On the other hand, these conditions are not just modeling reality, they are reality. They can therefore be considered to make for a realistic sample of conditions that you too may encounter in your own travels. The results and the corresponding conclusions, however, have to be treated with some care and the conclusions may be on the conservative side. In addition to this, the sample size is very small and does not result in statistically significant results.

The films were scanned at Amsterdam Schiphol Airport on all outbound flights and travelled around the world. Some of them stayed within Europe (Sweden, Italy), some were taken to the US or travelled through Germany to Japan and Korea. For security reasons the flight details are not disclosed.

Film stock

Table of contents

The tests were conducted using 120 roll film that I shot in a 645 format. Each film contains 15 frames of which: 1 identification frame that includes a description of the film and its identification number, 7 bracketed frames of a test scene of medium contrast (similar to the tests of I3A) in 1 EV increments from -3 EV to +3 EV, 4 blank frames to test for fogging of unexposed film, and 3 shots of the test scene at correct exposure.

The films used are Ilford FP4+ (EI 125), Ilford HP5+ (EI 400) and Ilford Delta 3200 (EI 3200). I chose these three film stocks, because 1) they can all be developed in the same recommended developer (Ilford DD-X), 2) their combination spans a wide range of often used box speeds and 3) are all part of my personal stock. I made the choice for black-and-white film, because I do not shoot color film myself and because the members of the I3A concerned themselves with color film already. Moreover, the I3A concludes that black-and-white film presents a worst-case scenario, and they recommend inspection by hand for every flight. The conclusions of this test, can therefore be considered to be on the safe side for color films too.

Although film stocks from other manufacturers, such as Kodak Tri-X/T-Max or Fuji Neopan Acros, may respond in slightly different ways, I do not consider these to influence the scope and validity of the final conclusions: the chemical and physical processes are the same. However, films are not always rated at their true ISO sensitivity, but rather at a box speed. This means that a film that would be ISO 200, could be rated as EI 400 for marketing purposes, because this will give good results in combination with certain developers. I shot all films at their box speed and used the manufacturer recommended developers and development times.

Because I did not want to break the bank, I had to limit the test to 6 rolls of FP4+, 6 rolls of HP5+ and 2 rolls of Delta 3200. Of each set, I picked one film which was used as a reference and was not exposed to x-rays. The remaining films are exposed between 2 and 11 times. To keep the time between the first exposure to x-rays and the last one to a minimum, I limited the field test to 11 scans. I consider this to be a realistic upper bound that may be experienced in real-life journeys. Most people will have their film scanned twice, or perhaps four times when a lay-over is necessary, and perhaps up to 10 times for extended travels.

The films were exposed on the 18th of May, 2016 and returned from the last flight on 4th of August, 2016. Due to hot room temperatures exceeding 24℃, the developing of the film was delayed until October 2016.  A room temperature close to 20℃ was considered necessary for developing at home, by hand. The films were kept together in the same location in the intermediate period and developed in batches of corresponding box speeds to guarantee that all films had undergone similar aging and can still be compared relative to each other.

Is 35 mm film shielded by its encasing?

Table of contents

The field test is conducted using roll film, which is not protected from x-rays by anything but the paper backing. On the other hand, 35 mm film comes contained in metal canisters. Do these metal containers make any difference for the final conclusions? To answer this question, a bit of physics is required. If you want to skip to the conclusions of this section, click here.

According to Ilford [1], the metal containers are made of 0.25 mm thick tin-free steel. This material specification leaves much to be guessed though, and I am not sure if technical support would actually know/share the actual composition. This “type of steel” is a multilayer material composed of the base steel, a film of metallic chromium, a film of chromium oxide and a coating of oil [2]. The relative thicknesses, however, can vary from manufacturer to manufacturer, and come to the specification of the buyer. And if that does not leave enough to be guessed already, the composition of the base steel is also unknown. To get a zeroth order approximation of the attenuation of the x-rays, I made the following assumptions:

  1. The influence of the thin oil film on the mass absorption is negligible.
  2. The chromium oxide is actually chromium (VI) oxide, also known as chromium trioxide (Cr2O3). This seems to be the oxide that is most used for plating purposes and is a product in the oxidation of stainless steel. This seems to be a good guess, given what we know about tin-free steel.
  3. The chromium oxide is applied in an average coating weight of 5 – 35 mg m-2 per side [2, 3]. It has a density of 2.7 g cm-3 and a molar mass of 99.99 g mol-1.
  4. The metallic chromium is applied in an average coating weight of 50 – 150 mg m-2 per side [2, 3]. It has a density of 7.19 g cm-3 and a molar mass of 52.00 g mol-1.
  5. The radiation propagates through the material as a plain wave, and internal reflections can be ignored.

From the average coating mass and the mass densities, it is easy to calculate that Cr2O3 is applied in a layer of 1.85 – 13 nm thickness. Likewise, the metallic Cr is applied in a 7 – 20 nm layer. The fact that the coatings are applied on both sides means that the total coating thickness adds up to less than 0.1 μm. In comparison to the total thickness of 250 μm, the effect of the mass absorption of these coatings is negligible and we can assume that we are dealing only with the base steel for this zeroth order calculation. This leaves us only to worry about the composition of the base material. Steels are composed of iron as the main ingredient, which is combined with either carbon or chromium as alloying metals for carbon steel or stainless steel, respectively. For sake of the argument, lets assume we are dealing with some kind of oversimplified stainless steel that is composed of 11% chromium and 89% iron by weight.

The intensity of the x-rays is attenuated by its interaction with matter. In this case, the intensity attenuation I/I_0 is given as a simple exponential function,

\frac{I}{I_0} = \exp\left( -\left( -\mu/\rho\right)\rho t\right),

where \mu/\rho is the effective mass attenuation coefficient, \rho is the mass density and t is the thickness. Using the data provided by NIST [4], it is easily found that at a scanner wavelength of 350 keV, chromium and iron have mass attenuation coefficients of 0.0097 cm2/g and 0.010 cm2/g, respectively. The effective value for the mixture, can then be calculated using a simple weighted average:

\left(\mu/\rho\right)_{steel} = \sum_i w_i\left(\mu/\rho\right)_i ,

where \left(mu/\rho\right)_i is the mass attenuation coefficient of species i and w_i is the corresponding weight fraction of the species in the mixture. In this case, this results in an effective mass attenuation coefficient of 0.0092 cm2/g.

As the rays have to pass through 0.25 mm of material, the total attenuation is less than 2%! In other terms, after 50 exposures, 120 film will receive a dose that is equivalent to 49 inspections for 35 mm film . In practice, this has no effect on the final conclusions.

Test scene

Table of contents

The subject for the test shots is shown in Figure 1. It is a scene of normal contrast and contains both white highlights, as well as deep blacks (the enlarger lens). In the left hand side of the frame, I have placed the test chart that I used before. The patches are used as an indicator of the densities that can be achieved with the specific film and my developing process.

Figure 1: Test scene.
Figure 1: Test scene. The scene is of normal contrast and features both highlights and shadow details. On the left is a series of black to gray squares and on the right is a set of white, gray and black cards.

The scene is lit using a flash that shoots into a reflective umbrella, as shown in Figure 2. The umbrella is placed such to illuminate the scene from a roughly 45 degree angle from the left and 45 degree angle from above. I metered the scene at f/8 and 1/60 s in the center of the scene (directly right of the smaller puppet sitting on the box camera) with the dome of the Sekonic L-758D light meter extended and pointed towards the camera. The meter reported a 100% flash-to-ambient illumination ratio. This means that the lighting is constant during the entire span of the shoot, even though the intensity of the ambient light changes over time. I repeated the metering before every roll to guarantee a consistent lighting. It remained constant to within the accuracy of the meter.

Figure 2: Lighting setup. A speedlight shoots intro a reflective, white umbrella and illuminates the scene from 45 degrees from above and 45 degrees from the left, give or take.
Figure 2: Lighting setup. A speedlight shoots intro a reflective, white umbrella and illuminates the scene from 45 degrees from above and 45 degrees from the left, give or take.

The f/8 aperture was chosen to sit right in the middle of the available aperture settings on the Bronica Zenzanon 75 mm f/2.8 PE lens that was used in this shoot. In this way, it allowed for up to 3 stops of over-exposure and up to 3 stops of under-exposure by simply changing the aperture, while the shutter speed could remain constant. When the illumination is dominated by the flash, the exposure can be considered to be fully determined by the aperture. The shutter time was the same for all frames.

For each film sensitivity, the power of the flash was changed to give the same exposure reading as before. The shutter speed and used apertures are the same as for all rolls.

Development

Table of contents

All films are developed in Ilford Ilfotec DD-X developer at a 1:4 dilution. In summer, the average room temperature in our apartment is close to 24°C, therefore I decided to bring the temperature of the chemistry to this level if it was not already and use the development times specified by Ilford, as cited in the last column of Table 1. I fully adhered to the manufacturer suggested developing procedure: the timer was started at the moment the liquid was poored into the tank at a quick rate, after which the agitation stick was gently twisted in alternating directions for 10 seconds to give the first film its first agitation. The tank was closed, and tapped firmly on the table top to dislodge air bubbles.  The tank was then given four inversions during the first 10 seconds of every subsequent minute. I started draining the developer in the last 10 seconds before the end of the cited development time.

Table 1: Development times provided by Ilford [15].
Stock Meter setting (EI) 20°C 24°C
Ilford FP4+ 125 10 min 8 min
Ilford HP5+ 400 9 min 7 min
Ilford Delta 3200 3200 9:30 min 7 min

The developer was freshly mixed for each roll and discarded afterwards. The acidic stop bath (Ilford Ilfostop, 1:19) and the fixer (Ilford Rapid Fixer, 1:4) were both freshly mixed and reused for each series (six rolls of FP4+, six rolls of HP5+, and the two rolls of Delta 3200). Both were used at room temperature, which was close to the developing temperature. The film remained in the stop bath for approximately 30 seconds under constant agitation using the swirling stick. This was followed by 4 minutes of fixer at 4 inversions at the first 10 seconds every minute. The film was subsequently rinsed in running water for 10 minutes, and treated with Kodak Photo-Flo 200 at 1:200 dilution for 1 minute and finally air dried until dry.

I kept the processing conditions as constant as I could. The FP4+ negatives, do look a little dense to my taste, so 8 minutes at 24°C is probably too much. The HP5+ negatives are properly developed and of normal contrast. The Delta 3200 rolls are both considerably fogged. This is very likely the result of an extended period at room temperature, while I was waiting for room temperature to drop.

Visual inspection and densitometry

Table of contents

From mere visual inspection of the negatives, it is clear that the x-rays have had an influence on the film and a clear darkening is especially obvious when the reference films are compared side-by-side to their siblings that were scanned 11 times (see Figure 3). Phototechnician Bram of the Color Utrecht photolab, told me the amount of darkening is not unlike that seen on older and/or slightly expired films. In practice he rarely notices any ill effects on print quality. The films used in this field test were all fresh, however. With the exception of the EI 400 film at 11 x-ray inspections, I found no banding in the negatives.

Figure 3: Blank frames of Ilford FP4+ film that was passed 0, 5 and 11 times through the x-ray scanners. A clear darkening of the negatives is visible. Photos were taken against a window on an overcast day. The visible structure is caused by the paper on which the negatives were mounted and are no artifact from the x-rays.

A transmission densitometer measures the fraction of light that is transmitted through a negative. The corresponding density of the negative is the logarithm (base 10) of the reciprocal of the transmission: a density of 1.0 means that 10% of the light was transmitted through the negative, while a density of 4.0 means that only 1/10.000 of the light was transmitted. By means of such a device, we can determine quantitatively the amount of darkening/fogging of the negative due to x-ray exposure.

Figure x: Barbieri Densy 511 at professional photolab Color Utrecht. This transmission densitometer was used to measure the density changes induced by x-ray exposures.
Figure 4: Barbieri Densy 511 at professional photolab Color Utrecht. This transmission densitometer was used to measure the density changes induced by x-ray exposures.

Color Utrecht allowed me to use their Barbieri Densy 511 densitometer (Figure 4) to measure the transmission density of the negatives. These were measured in five locations: on the unexposed frames of each rolls, black card, white card and both the left and right side of the background wall. The readability of this densitometer is 0.01 log D for each of the three channels of red, green and blue. Repeated measurements of the same spots of low and high negative density, yield a standard deviation of 0.002 log D and 0.001 log D, respectively, over 10 consecutive measurements. When the negative was removed and replaced between consecutive measurements and re-aligned to measure at (approximately) the same location, this increases to 0.01 and 0.004 for areas of low and high density, respectively.

Increases in base fog of unexposed material

Of each film a few frames were left blank and were not exposed to light. This allows monitoring of the base fog as a function of the number of x-ray inspections, storage conditions and development procedure. The changes in density with respect to the respective reference films are depicted in Figure 5.

Figure x: Measured changes in density (log10 D) of an unexposed negative as a consequence of x-ray inspections.
Figure 5: Measured changes in density (log D) of an unexposed negative as a consequence of x-ray inspections.

Up to five x-ray inspections, the changes in base fog remain within the reproducibility of the density measurement. The rolls of 9 and 11 x-ray inspections clearly show an increase in base fog. Perhaps a bit surprising, 11 inspections had less of an influence on the fog than 9 inspections did. One has to realize, however, that these rolls traveled through different airports (and thus different scans of different durations and at different orientations with respect to the x-ray source) and have been stored at different conditions during these journeys. The difference can also be explained by minor differences in the development, such as the developer temperature, which has a ±0.5°C uncertainty.

Changes in densities of exposed film

Exposed film is more sensitive to additional exposure, but can easily mask out minor differences in density. The changes in density (log D) are depicted in Figures 6 and 7 for Ilford FP4+ and Ilford HP5+, respectively. For each film, the density of the highlights (represented by the white card), the shadows (represented by the black card, and light midtones (represented by the wall) were measured.

Figure 5: Changes in density for Ilford FP4+ as a function of the number of x-ray inspection.
Figure 6: Changes in density for Ilford FP4+ as a function of the number of x-ray inspection. The lines are included for better legibility and do not signify linear interpolation.

Surprisingly, the changes in density are already apparent after two inspections and no clear trend is visible in the results for FP4+. Variations of roughly ±0.10 log D are recorded irrespective of the number of x-ray inspections.

Figure 6: Changes in density for Ilford HP5+ as a function of the number of x-ray inspections. The lines are included for better legibility and do not signify linear interpolation.
Figure 7: Changes in density for Ilford HP5+ as a function of the number of x-ray inspections. The lines are included for better legibility and do not signify linear interpolation.

The results of the more sensitive HP5+ films are strange by comparison. A downward trend is visible in all four measurement locations, whereas a density increase is expected. So far I have been unable to arrive at any reasonable explanation for this trend. The absolute differences are, however, within the same bounds as found in the FP4+ films. With the exception of the datum for 2 x-ray inspections as measured on the location “wall (right)”, the overall measured relative changes are correlated with the number of scans. In other words, all four measurement location exhibit very similar changes in density with respect to the other datums: if the shadows go up, so do the highlights and the midtones. This is visible in the results for both FP4+ and HP5+ films. This supports the hypothesis that the main effect of the x-ray inspections is uniform fogging of the film.

Any significant fogging would affect the shadows more than the highlights. On the print, this will become evident as a lower contrast as the ratio between shadows and highlights would become smaller. Figure 8 depicts the ratio of the measured densities as measured on the black card (“D Max”) and the white card (“D Min”). The names refer here to the maximum and minimum measured density values, respectively, and not to the more conventional base fog level and maximum attainable density.

dr
Figure 8: The ratio between the densities measured at the white card and at the black card.

As is obvious from the graph, the changes in this contrast are small and can reasonably be attributed to variations in the development conditions such as developer temperature or concentration.

Printing

Table of contents
To determine whether the measured changes in density pose any problem in printing, I made work prints of the FP4+ series for a frame of normal exposure of the reference roll and the roll that was exposed to 9 x-ray inspections. The results are shown below in Figure 9. Both prints were exposed under the enlarger for the same exposure time of 24 seconds on Ilford MG IV RC paper with a pearl finish and an Ilford grade 2 under-the-lens filter. The prints were developed in fresh Ilford Multigrade developer (1:9 dilution) for (60±3) seconds at room temperature.

Figure 9: Photographic reproduction of prints of the reference scene, both shot under identical lighting conditions on Ilford FP4+, rated at EI 125. Prints were made on Ilford MG IV RC paper with a pearl finish. Top: 9 exposures to x-rays; Bottom: 0 exposures to x-rays. Click on the image for the full sized version.

To the unaided eye, the prints are nearly indistinguishable in terms of tonality and contrast (there is of course the awkward difference in crop due to the misplaced lamellae on the negative carrier). The photographic reproduction (unaltered except for a change in white balance and crop) exaggerates the difference slightly in my opinion. The differences are so small, that I didn’t feel the need to compensate the exposure time for this test. A minor adjustment in the exposure time would eliminate the difference.

Conclusions

Table of contents

A simplified theoretical study of the response of silver halide emulsions to x-ray exposures gives good cause for photographers to suspect x-ray scanners from potentially degrading image quality, by adding additional fog or banding in the images. Systematic tests conducted by the International Imaging Industry Association (I3A) in the late 1990s and early 2000s confirm that this was indeed a realistic concern for high speed colour negative and colour reversal film. Although their results have served as a good benchmark for the industry for many years, they cannot be considered to be representative of modern x-ray scanners and they only briefly touched on the results for black-and-white films.

This field test was conducted using Ilford FP4+, Ilford HP5+ and Ilford Delta 3200, which were respectively rated at EI 125, EI 400 and EI 3200. Films were passed up to 11 times through airport x-ray scanners while it remained packed in regular hand luggage during routine luggage inspections on international flights in Europe, the US and Asia. For this specific instance, no image degradation was observed in the final printed images. Visual inspection of blank frames (that were not exposed to light, but were exposed to x-rays) show a clear fogging of the film. Densitometry at several locations on the respective reference photos, however, show no trend indicative of incremental image degradation with increasing number of scans. The results support the conclusion that the effects are limited to mild fogging of the negative. Surprisingly, no significant differences were found in absolute density changes in relation to film sensitivity. Moreover, I found no distinct banding in the images that was apparent to the unaided eye.

The lack of significant statistics do not allow us to draw general conclusions or estimate the risk of unrecoverable damage by x-ray inspection. I do consider the test to be representative of real traveling conditions, and representative of what many of you will experience in practice. This does not mean that your films are completely safe from any harm. However, the results are in line with the I3A reports, which makes me confident that an upper limit of five x-ray inspections can be deemed safe and will not result in (unrecoverable) image degradation in practice. However, it is wise to always kindly request a manual inspection to minimize the exposure to x-ray radiation. If the opportunity arises, get your film developed locally to limit the number of x-ray inspections that may harm the latent, undeveloped image.

Summary

Table of contents

The dreadful horrors of flying with film start already before you even enter the plane: the security check and the obligatory pass of hand luggage through the x-ray machine. There is no denying it: x-rays do result in a latent image on the negative. There are, however, two major differences with the effects of visible light.

  1. The high photon energies associated with x-rays makes them set off a chain reaction in the emulsion that is 10.000 times more potent in forming an image than visible light.
  2. This reaction is so energetic, that the image formation is scattered over a large area and can span multiple grains. This generally manifests itself in a fogging of the image. For visible light on the other hand, the silver forms close to where the photon struck and stays within the boundaries of the grain. This yields a distinct image, which we consider a photograph.

Because an even fogging of the film can be ‘printed through’ and corrected for in printing by increasing the exposure time or paper gradation, it is generally not considered a severe problem. High intensity beams that are employed in some x-ray machines can form distinct bands in the imagery. These are not always clearly visible, because they are obscured by the actual image. For this reason, airport security will tell you that the scanners are safe for conventional sensitivities up to EI 1600. The International Imaging Industry Association (I3A) tested the x-ray machines of the late 1990s/early 2000s for the FAA, and found that for colour negative film this is indeed the case. For black and white film the results seemed less favorable, but only very little information was published.

The field test was conducted using Ilford FP4+, Ilford HP5+ and Ilford Delta 3200 film which were shot on box speed. The films were exposed to either 0, 2, 4, 5, 9, or 11 exposures by airport x-ray scanners (Delta 3200 only 0 or 2 exposures) and developed in Ilford Ilfotec DD-X. Visible inspection shows noticeable fogging for EI 125 film (box speed) at 5 passes through the scanner. Densitometry shows a difference in density of most ±0.12 log D with respect to the corresponding reference shots. The measured changes in density do not monotonically vary with the number of x-ray inspections, which hints that various factors are at play. Different storage conditions and small deviations in the development process can cause similar differences. Interestingly, no clear differences were found between films of different box speeds.

To determine whether the measured changes in density pose any problem in printing, work prints were made of the FP4+ series for a frame of normal exposure of the reference roll and the roll that was exposed to 9 x-ray inspections. At identical settings of the enlargers, the prints come out nearly identical and are hard to discern by the unaided eye.

The small number of samples and uncontrolled conditions make it impossible to give hard conclusions and recommendations regarding flying with film and the inevitable x-ray inspections. The results, however, are in line with prior findings of the I3A and when you maintain a maximum of five x-ray inspections the affects should go unnoticed. This is not to say that damage may not occur on your specific journeys. You may encounter a combination of bad conditions such as a warm climate, an old or high intensity x-ray inspection system, or for example a prolonged inspection, which may all result in image degradation. In case your images are irreplaceable, I advice to get your film developed locally and always kindly request a manual inspection.

Disclaimer

I do not accept any liability regarding any damage done to your films by x-ray exposures or otherwise. These tests were conducted to satisfy my own curiosity and supply supporting information regarding the underlying physics, chemistry and considerations. You and only you are responsible for any damages incurred by x-ray inspection of your films. If the imagery is irreplaceable, insist on a hand inspection or take relevant measures to prevent such damages.

Acknowledgements

I would like to thank dr. Luigi Sasso, Long Wu, dr. Laura Iapichino, Rob Eling, Svenja Woicke, and dr. Alejandro Aragón for taking films with them in their hand luggage. Without them, this field test would have been impossible within any reasonable time. In addition I would like to thank Patrick and Bram of the Color photo lab in Utrecht, the Netherlands for making their densitometer available to me. Without that, the entire experiment would have been limited to mere qualitative comparisons.

Cover image
The original uploader was IDuke at English Wikipedia (Transferred from en.wikipedia to Commons.) [CC BY 2.5 (http://creativecommons.org/licenses/by/2.5)], via Wikimedia Commons

References

[1] Private e-mail conversation with Ilford Photo’s Technical Support on April 26, 2016.

[2] JFE Steel, “TIN PLATE and TIN FREE STEEL”, JFE Steel Catalog, Retrieved on 1 May 2016 from http://www.jfe-steel.co.jp/en/products/sheets/catalog/b1e-006.pdf

[3] “Tin Free Steel”, Ispat Guru, Retrieved on 1 May 2016 from http://ispatguru.com/tin-free-steel/

[4] NIST, “X-Ray Mass Attenuation Coefficients”, NIST Website, Retrieved on 1 May 2016 from http://physics.nist.gov/PhysRefData/XrayMassCoef/tab3.html

[5] Eastman Kodak Company, “Baggage X-ray Effects on Film”, Eastman Kodak website, Retrieved on 3 May 2016 from http://www.kodak.com/global/en/service/tib/tib5201.shtml

[6] Fuji Film, “Is there any danger to bringing my film through an airport x-ray machine?”, Fuji Film Website, Retrieved on 3 May 2016 from http://www.fujifilmusa.com/support/ServiceSupportProductContent.do?dbid=670359&prodcat=238119&sscucatid=664277

[7] User bmattock, “How many X-ray scannings have you gone through without problems?”, the rangefinderforum.com (11-29-2009, 08:29), Retrieved on 3 May 2016 from http://www.rangefinderforum.com/forums/archive/index.php/t-82810.html

[8] The I3A Committee for Integrity in Transportation of Imaging Products, “Study Of Damage To Photographic Films X-rayed by the Rapidscan Security Checkpoint Scanner” (unknown), Accessed on 10 May 2016 using the Internet Archive.org WayBackMachine, which archived the document  on 16 March 2007. Retrieved from http://web.archive.org/web/20070316121548/http://www.i3a.org/pdf/I3AFilmXrayTestReport.pdf

[9] J. Auer, G. Barr, G. Barrera, P. Bovee, T. Dufficy, L. Forsmark, S. Howe, M. Kahn, J Pardo, R. Parsons, J. Placko, J. Roberts and P. Roth, “The CTX-5000SP and Camera Films” (unknown), Accessed on 10 May 2016 using the Internet Archive.org WayBackMachine, which archived the document  on 16 March 2007. Retrieved from http://web.archive.org/web/20070316121453/http://www.i3a.org/pdf/ctx-5000sp-camera_films_report.pdf

[10] R.W. Gurney and N.F. Mott, “The theory of the Photolysis of Silver Bromide and the Photographic Latent Image”, Proceedings of the Royal Society, 164A, 151-67 (1938).

[11] J.F. Hamilton, “The silver halide photographic process”, Advances in Physics, 37 (4), 359-441, Online: http://dx.doi.org/10.1080/00018738800101399

[12] H. Hoerlin and F.A. Hamm, “Electron Microscopical Studies of the Latent Image Obtained by Exposures to Alpha Particles, X-Rays and Light”, Journal of Applied Physics24 (12), 1514-1519 (1953).

[13] A. Charlesby, “The action of electrons and X rays on photographic emulsions”, Proceedings of the Physical Society, 52 (5), 657-700 (1940).

[14] H. Hoerlin, “The Photographic Action of X-Rays in the 1.3 to 0.01 Å Range”, Journal of the Optical Society of America39 (11), 891-897 (1949).

[15] Ilford, “Technical Information: Ilfotec DD-X, Ilfotec LC29 and Ilfosol S Film Developers”, Ilford Website (unknown), Accessed on 13 August 2016. Available on http://www.ilfordphoto.com/Webfiles/2011427124733149.pdf.

Back to Top