It is a property of ABS and other plastics that they will undergo photooxidative degradation – the materials will yellow with age, particularly in response to ultraviolet light.

In 2025, the LEGO Group introduced a new white pigment in selected sets. This colour, referred to as 426 White V3 is more opaque than the 1 White that has otherwise been in use since LEGO® Bricks started to be coloured by adding pigment to raw ABS stock in the early 2000s. Given the relative opacity to 1White, I wondered if there might be a difference in the yellowing exhibited by these elements after prolonged exposure to sunlight. So I masked off some tiles,

I will apologize for drifting back into high school science practice write-ups and the lack an a priori power analysis, and subsequent dodgy statistics, as we set out to answer the question “Is the new LEGO pigment 426 White V3 less susceptible to yellowing when exposed to sunlight?”

Method

I mounted a 2×4 tile of both 426 white and 1 white on a plate, and masked the lower half of them with a black tile and some black duct tape. I taped the assembly to a post in my backyard in suburban Melbourne (37.8117º South; 145.1ºEast) , and exposed the assembly to the elements for the better part of 12 months (technically, 51 weeks). I placed the models that I was using as a control back in their boxes, and stored them inside.

Due to difficulties in reliably reproducing photographic conditions, as well as being distracted by important and necessary aspects of life, my initial intention to compare the specimens on a monthly basis fell by the wayside.

At the end of the experimental period, I removed the plate from the verandah, and removed the duct tape and overlying tiles.  I gave the sun-exposed tiles a wash with soap and water to remove any residual dirt that had settled on the tiles.

I retrieved the 10360 Shuttle Carrier Aircraft and the Creator 31117 Space Shuttle Adventure from storage. These sets were the source for the plates I used for the experiments. Both sets had been stored inside, in their cardboard boxes during this time.

I removed an additional white tile from the wings of both, to use as a ‘non-exposed control tile’ and mounted them next to the experimental tiles.

The tiles were inspected and photographed. A colour picker tool (Adobe Express) was used to identify RGB channel values at random points in each of the six areas being examined, and a yellowing proxy was used to determine the degree of yellowing seen.

No attempts were made to assess the relative strength or brittleness of the sun-exposed elements.

Results

Qualitative observations:

On initial inspection, the sun-exposed area of each 2×4 tile had changed colour significantly compared to both the masked and control areas. Interestingly, there are also faint radial ‘crazing’ patterns on the surface of each tile, originating in different areas. This pattern is not visible on the non-yellowed part of the tiles, and persisted after scrubbing with soap and water.

This pattern is more visible after adjusting the contrast, blackpoint and clarity settings for the image.

The new white pigment maintained its relative opacity to the older one as seen by placing a light bar behind them

Quantitative analysis

I photographed the tiles and used an RGB Yellow proxy to establish the relative degree of yellowing in each test sample against the equivalent control. Bricknerd published an article a few years ago looking at the effects of light on LEGO elements, including white elements, examining the way that different RGB channels varied as the samples were exposed to light.

In the absence of a spectrophotometer, I have adapted this technique and used a ‘Yellow proxy’ to assess the colour change seen in the samples.

What’s a yellow proxy? As far as light is concerned, Yellow is a secondary colour, made up of red and green components. Using the RBG Colour model, white and greys have a cluster of similar numbers 0,0,0 is black; 255,255,255 is bright white, while 128,128,128 will be somewhere in between.

When we scan over the picture for a point’s RGB values, we get values for red, green and blue channels respectively. An increased yellowing will see an increase in the red or green channels and/or a relative drop in the blue over time. We can calculate a ‘yellow proxy’ = red channel – blue channel.

I took a single picture of the experimental elements (masked and exposed regions) as well as a control tile. I fixed my camera’s white balance at 5000K and turned off the overhead light: all lighting was natural sunlight, diffused through the corflute walls of a Foldio4 Lightbox. Blinds were open and it was a sunny day.

I imported the RAW photograph into Adobe Lightroom. As the intention is to evaluate yellow deviation from the control element, the white balance was set to correspond with the control section for 1 White. The image was then opened in Photoshop, where a colour sampling tool was used to return RGB values averaged over a 5×5 pixel range. Values were selected from three random points in each region: Control-old white; test element – old white, masked; test element old white, exposed; control-new white; test element-new white, masked; test element-new white, exposed.

For each of these colour points, I calculated a yellow proxy, and the average value for each region was calculated.

Yellow proxy = red component – blue component.

Both the 1 White and 426 White V3 demonstrated a similar degree of yellowing using this form of yellow proxy. The mean for each RGB point was calculated, and plotted on a seperate colour wheel for 1 White and 426 white V3.

The control and masked regions were remarkably consistent between different testing regions. (range of yellow proxy = 0-1 for old white; 1 for 426 white).

The sun-exposed test areas both showed a change in mean yellow proxy of 20.33 for the 1 White, while we measured 20.33 for 426 White V3. The total range was similar for both: 19-21 vs 19-22

The mean values for 1White control, 1 White sun exposed and 426 White, sun-exposed were plotted on a colour wheel, showing a slight difference between the 1 and 426 exposed regions:

I don’t know for certain I have enough sample points to establish whether there is a significant difference between the two yellowed regions. My feeling is that we might need dozens of datapoints in each group to establish whether this is, in fact, a small but significant difference. Ideally, we would collect dozens of data points, from multiple copies of the same element. Note to self: next time, consider a power analysis and a statistician, to ensure the result is appropriately robust,

Discussion

This study was set up to create the maximal sun damage to the LEGO elements as was possible, by removing all forms of artificial barriers and maximising exposure to the UV light, short of carting the bricks into the middle of the desert. As such, the changes observed will be much greater than those seen while having a set on display, in a relatively darkened room, in Northern Europe.

When I set the experiment last year, I anticipated monthly progress shots. Given the difficulty of maintaining the same quality of lighting over that period, I only used photos from the end of the experimental period. that and I might have forgotten, somewhere along the line.

In the past, yellowing of LEGO Bricks has been attributed to the use of bromine used within the plastic to act as a flame retardant. This has not been in general use for this purpose since the turn of the century. The specific stabilizers incorporated in contemporary bricks have not been freely named in recent years.

However, yellowing on exposure to UV light is a property of plastics – a consequence of the molecular bonds in the polymer breaking down in response to prolonged exposure to heat, chemicals or radiation, resulting in oxidation of the polymer subunits. Photodegradation happens independently of the colour of the plastics concerned, but it is far more apparent with white plastics.

Ultimately, the degree of photodegradation observed was fairly similar between the two white pigments, suggesting that the yellowing is a property of the ABS, and no advantage or disadvantage is conveyed by the nature of the white pigments tested.

Environmental conditions:

As I mentioned, we were aiming for a ‘maximal effect’ experiment – not ‘everyday use’. What sort of conditions were the tiles exposed to?

Over the previous 12 months, temperatures in Melbourne dropped as low as 1.6ºC, and rose to 42.9ºC (34ºF to 109.2º), with an average minimum 11.4ºC; average maximum 20.6º C over the study period. (Bureau of Meteorology, Melbourne)

The UV index in Melbourne varies across the year, and as you can see, Copenhagen has a significantly lower UV index throughout the year compared with Melbourne. As such, you would expect a greater degree of yellowing on exposure to the southern Australian sun, compared to northern Europe.

That said, the annual hole in the ozone layer over Antarctica in spring 2025 was one of the smallest since 2009. The current research suggests that this will be healed up by 2060.

Bear in mind that the result that we see is the result of environmental exposure of LEGO Elements, not just to sunlight but also wind, atmospheric pollutants and the occasional rogue hail stone. Perhaps the grazed linear patterns seen on the exposed surfaces of the tiles is a sign of greater damage done to the plastic than simply colour changes.

Work by other investigators

Other content creators have performed extended sunlight exposure tests on LEGO elements, including Town Creator and Epikbricks.

Town Creator looked at changes to the ‘Everyone is awesome’ set – one stored in a dark room, another within an area where it gets exposed to natural light for part of the day. While minimal yellowing is apparent to the viewer – perhaps the orange bricks demonstrated the strongest subjective color changes. He also reported a subjective change in the clutch power of the sun-exposed elements. Town Creator is based in Norway, and the UV index is typically even lower in Norway than in Denmark.

Epikbrick’s primary experiment involved progressive masking of a white baseplate over a 12 month period. He demonstrated variable yellowing in a white baseplate – baseplates are made of high-impact polystyrene, and might degrade in a different way than ABS elements.

I find myself wondering about the role of other barrier techniques: UV window films and acrylic cases. Standard Acryilic blocks UV-B and UV-Crays, but does not have a significant effects on UV-A. Additives to the these plastics such as benzophenones or benzotriazoles can reduce the UV-A penetration as well. Something to consider if you are investing heavily in plastic cases for your collectibles.

Having established that yellowing with 426White V3 is essentially the same as that seen with 1 White, I find myself asking questions that I had not been particularly concerned with over the years, and could be a launching pad for future studies:

• Does the yellow proxy technique work consistently when investigating pigments other than white?
• How does ‘yellowing’ manifest in transparent MABS elements? Previous transparent elements were made of polycarbonate, which is said to refelect/protect against UV radiation. We would expect MABS to behave somewhat differently.
• How does the degree of yellowing with current LEGO Elements compare with those that incorporated bromine in the plastic as a flame retardant?
• Do UV filter window treatments slow or prevent photodegradation resulting from sunlight driven UV exposure?
• Is there a difference in the way these pigments respond to using hydrogen peroxide and daylight to reverse the yellowing?

Conclusion

In conclusion: After 12 months exposed to the weather and Australian sun, 426 White V3 underwent a similar change in colour as the standard 1White.

This is a negative result compared to our starting hypothesis, but is just as valid in contributing to things we know about this new pigment. We cannot exclude exposure to other atmospheric contaminants from potentially causing some of these changes.

There is scope for further research to validate the ‘yellow proxy’ as a tool to assess photodegradation after exposure to UV light. This technique requires further development and validation across different LEGO pigments.

The specific origin and nature of the new linear/radial patterns on the exposed tiles remains unclear.

There is also scope for controlled trials to assess the effectiveness of UV-protecting window films and acrylic display cases to minimise sun damage to LEGO elements.

If you are considering

Conflicts of Interest

The author is a member of the LEGO Ambassadors Network and has received review sets from the LEGO Group to create online and offline content. Elements used in this experiment were taken from sets provided by the LEGO Group for review purposes. All opinions and ideas are independent of this support.

If you have enjoyed this post, please feel free to share it with your friends, families and communities. You can also keep up to date with us by following The Rambling Brick on our socials: Facebook, Threads, Bluesky, Tumblr (or just sign up for our mailing list) and find some extra content on Instagram and TikTok.

We really value your feedback: please let us know what you think about the sets we review, and the articles we post. And if you wish to support the Rambling Brick, consider clicking on our affiliate links before going shopping at LEGO.com. The Rambling Brick receives a small commission, and it costs you nothing extra.

Until Next Time,

Play Well!