Neutral Gray 43 ~ The Clue is in the Name

Something of a departure from regular fare - again - as a discussion at Britmodeller about the best hobby paint to represent the USAAC paint colour Neutral Gray 43 (NG43) triggered some thoughts on the matter. Regular readers will know that one of the motivations behind this blog is an attempt to separate colour facts from colour fictions - or perhaps more correctly from colour 'myths'.  Not necessarily in any kind of revisionist or heretical way, although that often seems to be the result, but by examining what is actually known, as dispassionately and objectively as is possible and applying established colour science to it. This is more often than not contrary to perpetuating what is thought to be known and which, when investigated, is sometimes found to be based on unreliable or even obscure foundations. 

The first reality check with NG43 is the idea it that it was a "pure" neutral grey consisting only of black and white pigments. It wasn't. The pigments specified for NG43 in 14105 (camouflage pigmented nitrate lacquer) and 14106 (camouflage pigmented nitrate dope) were titanium dioxide, yellow iron oxide and lamp black. More on these pigments later.

The second reality check is that there is only one colour value in FS 595B that approximately matches these pigments (see below) and none that match the colour. And this is further borne out by conducting colour comparisons to crunch the numbers. There is not a single FS value that comes even close to the 2.0 or less requirement of the DE2000 difference calculation criteria.  The FS values most often cited for NG43, FS 36270 (supposedly in error) and 36173 (supposedly correct)  both contain rutile titanium dioxide (white), phthalocyanine blue (red shade) and carbon black (blue shade). In addition 36270 contains benzimidazolone yellow whilst 36173 contains red iron oxide (blue shade). This is not surprising as contrary to what is often asserted NG43 did not evolve into any other standard colour after its appearance in Bulletin 41 of Sep 16 1940. It was replaced by the ANA paint colour Sea Gray 603 introduced in ANA 157 on Sep 28 1943 and that evolved to become FS 36118.  

The closest FS value to NG43 is FS 36152 @ 3.54. However, it is still more blueish in appearance and the constituent pigments are rutile titanium dioxide, phthalocyanine green (blue shade), chrome yellow (red shade) and carbon black (blue shade). Chrome yellow is a stronger, brighter and clearer yellow pigment than yellow oxide. The closest FS value to NG43 in pigment content terms is FS 16492 which contains almost exactly similar pigments of rutile titanium dioxide, yellow iron oxide and carbon black but it is clearly much lighter than NG43 as a result of ratio differences.

I do not know where the idea of a blueish NG43 comes from, but there is a prevalent tendency for "greys" to be perceived as blueish and many of the NG43 hobby paints seem to show this trait. Even though many modellers repeat the popular idea that NG43 was a pure mix of black and white they seem comfortable with accepting these blueish grey hobby paints as representative and in citing FS 36173 as a match!  Perhaps this operates on the "decorator principle"? I also get the impression that a lot of these matches are being made under artificial hobby room tungsten light which draws 36173 closer to a warmer, neutral grey. Certainly the swatch included in J F Dial's 1964 'United States Camouflage WW II' (Scale Reproductions) is significantly more blueish in chroma than the paint chip and printed chip in both the Archer books*. Again I emphasise that the discussion here concerns the paint colour standard and does not necessarily mean that all applied paints matched that standard perfectly and/or did not exhibit colour shift as a result of manufacturing processes or environmental and age related stresses. Please, no more colour photographs to "prove" points about paint!

Another clue to the real characteristics of the colour is the fact that for the first time in Spec 14057-C of 27 Dec 1939 Munsell colour notations had been cited for the colours and Neutral Gray 32 was notated as Munsell N5 (in the 1929 system). Neutral Gray 32, an earlier temporary camouflage paint, was identical to the subsequent NG43. The 1929 Munsell N5 is entirely consistent with the Archer chips.

Both Testors (#1725) and Xtracolor (X133) versions of Neutral Gray are said to be matched to FS 36270 and 16270 respectively, but the Testors paint is actually manufactured only with titanium dioxide (15-20%) and carbon black (0.1-1.0%), although it contains hydrous aluminium silicate (an extender also known as China clay) and magnesium silicate (a filler also known as 'talc'). I haven't tested the Testors but I presume it to be a genuine neutral gray in colour? The Xtracolor paint appears as a very close match to FS 26270 but is a little lighter and warmer. White Ensign Models (WEM) Colourcoats ACUS13 WW2 USAAF Neutral Gray has the hybrid designation "ANA 43' (NG43 was never a ANA paint colour). It is a tad darker and slightly more blue-green in hue than FS 36173.

Titanium dioxide (C.I. pigment white 6) was developed in the 1920's and was in wide scale use as a white pigment by the 1950's. Prior to its use white pigments used in paints were compounds such as lead carbonate, zinc oxide, antimony oxide, lead silicate, lead sulfate and co-precipitated forms of these compounds. Other than zinc oxide most of these other forms of white pigment have become obsolete, not least because of issues of toxicity. It is manufactured in two processes, by sulfate and by chloride from ores such as ilmenite or rutile. In the sulfate process the ilmenite ore is reacted with sulfuric acid to produce titanyl sulfate, which is then hydrolised to produce hydrated titanium dioxide. This in turn is then calcinated to produce titanium dioxide pigment. There are two main forms - rutile, with resistance to chalking - and anatase, with a tendency to chalk significantly.

As a digression, another reality check for modellers is the prevalent idea that only the sun degrades paint and then only to "fade" it. In the exposure environment the chemical and physical processes affecting a paint surface are complex, combined and cumulative as it is subjected to diverse and variable environmental stresses. The most significant are heat, oxygen, sunlight, water (both in liquid and vapour form), mechanical and thermal stresses, atmospheric pollutants and various chemicals which might be applied to it, usually deliberately to clean it or unintentionally from materials or solvents associated with other servicing or maintenance activity. The paint film also contains decreasing levels of residual solvent or water, or both, retained after application, as well as pigments, fillers, additives or trace metal impurities incorporated during manufacture - all of which will probably have some photochemical activity. The physical effects that can be expected include gloss loss, colour change, dirt retention, chalking, cracking, delamination, blistering, fouling and corrosion. 

The typical effects of chalking result in a whiteish-grey powdery patina across the surface which is not that obvious to the eye unless you rub your hand over it. Titanium dioxide is both a UV-activated oxidation catalyst and a UV absorber, with a tendency to absorb light at the blue end of the spectrum and to appear as a yellow toned white. As an oxidation catalyst free radicals are formed on the surface and oxidise the binder by photocatalytic degradation. This reduces any gloss and produces a friable layer on the surface of the paint film. The process will be exacerbated under direct UV exposure but even surfaces out of direct light will not be immune, for example a combination of extremes of heat and humidity will cause a similar degradation. Any paint that contains a majority of the anatase form titanium dioxide (such as greys and light blues) would tend to exhibit more chalking, which is why Ocean Grey, say, can appear to "fade" much lighter than adjoining Dark Green. A common fallacy of modellers is to assume that all camouflage paint colours "fade" or degrade at the same rate.

The effect of chalking on a grey paint would be to make it appear lighter (and with other colours lighter and more greyish) but typically the surface would be variegated rather than even in appearance, especially where it has been subjected to different degrees of exposure. How much lighter the grey becomes depends on the duration of exposure and degree of degradation, reaching a point where it could be almost off-white in appearance and ultimately where the paint breaks down through the whole film and begins to reveal the primer. Regular maintenance treatment will reduce the chalking process or delay it but not eradicate it entirely.

Now, to return to NG43, what should make us think that the form of titanium dioxide used in the paint was anatase?  Well, the same specifications make a point of requiring the "anti-chalking" variety of titanium dioxide for the Insignia White and Insignia Blue paints, a requirement omitted for NG43, which strongly suggests (and no more than that) that NG43 could be manufactured using the anatase form of the pigment.

Yellow oxide - or yellow iron oxide - exists in synthetic (CI pigment yellow 42) and natural forms (CI pigment yellow 43) It is produced synthetically by hydrating ferric oxide to a range of dull reddish-yellow to yellowish-orange brown hues, the final colour determined by the size of the particles in turn controlled by rate and duration of oxidisation and the concentration and pH of solutions. Yellow iron oxide is not therefore an identifiable colour by name alone. It is a relatively cheap pigment used extensively to produce cream tints with titanium dioxide. In common with red oxides it provides good light fastness (colour retention) and heat resistance, absorbing UV radiation to an extent which gives good durability. It has good opacity and good chemical resistance. It has poor tinting strength which means that when added to, say, titanium dioxide, a lot goes a very little way and the shades produced by it are usually dull rather than bright.  Why was it added to NG43?  Possibly to counter the chalking tendency of the titanium dioxide or to counter a blueing tint from the use of blue shade carbon black. The effect of a titanium dioxide and yellow oxide combination would certainly produce a warm tint grey rather than a cool blue-grey. 

Lamp black (CI pigment black 7), not to be confused with carbon black (pigment black 6) although related and sharing the same CI # 77266, is commonly used by the paint industry to produce a superior black colour. Carbon Black is much better in colour and tinctorial value than Lamp Black but the difference between their structures is not completely understood. Carbon blacks are often known by the name of the hydrocarbon they are made from, e.g. Acetylene Black or Benzol Black. It is usually made by heating oil, gas or acetylene at very high temperatures with limited air.   It has a very small particle size which can cause dispersion problems leading to poor gloss and poor colour. It can also have stability issues in air drying media and when combined with other pigments such as titanium dioxide it can flocculate, changing colour in storage. In pigment form, far from being constant, the hue varies from deep brownish or bluish or greyish to deep black.

* The Official Monogram US Army Air Service & Air Corps Aicraft Color Guide, Vol.1 by Robert D Archer (Monogram Aviaton Publications, 1995) and USAAF Aircraft Markings and Camouflage 1941-1947 by Robert  D and Victor G Archer (Schiffer Publishing 1997).