Return to Sherwin Family Enterprises Homepage
Spectral Analysis of Various Papers and Paints

As a Retired with 40 years plus experience as an Industrial Research and Development engineer turned watercolor artist, I am obsessively interested in how the things I use work. I was curious about obtaining objective measurements of the color of the paints and papers I am using.

The color of paints and paper can be objectively measured with a device called a reflective spectrometer. That is why I decided to build such an instrument. Once the spectrometer was fully funcional and wavelength calibrated, I measured the spectra of various watercolor papers, shiny aluminum foil, and various white through gray to black paints, and also various colored paints prepared from pigments from both Natural Earth Paint and my own wild gathered and home processed pigments.

The 45/45 Degree Reflective Spectrometer, like the one I built, objectively measures the reflected light of a "white" light source from a sample surface set at a 45 degree angle relative to the direction the light source is shining from, and measured at the adjacent 45 degree angle. (A 90 degree difference.)

Upon reporting the results of my study to Natural Earth Paint and studying the results myself, I thought that other watercolorists might find the report of value. So I set up this web page and pasted a link to it in "All About Watercolor"


Figure 1. shows my home made spectrometer with a chip of "Arches Aquarele Cold Press Paper" painted with NEP Emerald Green paint in the sample holder.

Figure 1.



Figure 2. shows the spectrometer with the white light source attached.

Figure 2.



Figure 3. shows the sample end of the spectrometer with the light source moved out of position to show the entrance slit of the spectrometer .

Figure 3.



Figure 4. shows the color swath taken from the spectromerers camera. The columns of the color swath image for each paint sample were summed to generate the curves for each respective sample.

Figure 4.



Figure 5. shows the calibration to Red-(550-nm), Green-(522-nm) and Blue-(480-nm) LED light sources.

Figure 5.



All paint data is taken on watercolor swatches painted on Arches Cold Press Block paper. I haven't, at this point, come with a standard, uniform method of color swath preparation. I try to apply two uniform layers of each paint on its respective chip, but I am not currently satisfied with the uniformity of all the paint chips except for a couple of successful chips


Figure 6 shows the colored paint swatches that I used for the colored paint portion of this study.

In order left to right, top to bottom, they are:

1-1 NEP-French-Torquoise, 1-2 NEP-Emerald-Green, 1-3 NEP-Mayan-Red, 1-4 SFE-Limestone-Gray,
2-1 NEP-Ultramarine-Blue, 2-2 NEP-Bright-Yellow, 2-3 NEP-Venetian-Red, 2-4 SFE-Sandstone-Tan,
3-1 NEP-Ultramarine-Purple, 3-2 NEP-Yellow-Ochre, 3-3 SFE-Terracota Orange, 3-4 SFE-Red-Sandstone-Brown.
Figure 6.



In the following figures / plots, the wavelength regions are indicated by the multi color line in each plot. They are in order, left to right, Infrared-(Black), Red-(Red), Orange-(Orange), Yellow-(Yellow), Green-(Green), Blue-(Blue), Purple-(Indigo), UltraViolet-(Black).

All the following curves , except number 8, use the same data and display it in different ways to enable more understandable analysis.

NEP- prefixes refer to watercolor paints processed from Natural Earth Paint pigments. SFE- refers to paints produced from pigments that I myself procured, either by purchasing the raw materials or by gathering them in the wild.


Figure 7 shows the raw sensor response measured maximum white through various whites through blacks to sensor minimum background signal.

Figure 7.



Figure 8 shows the data normalized to "SFE-Dolar Tree Marble White". Thus this paint appears as the cyan line at Reflective Response = 1.0. TheDark Reference refers to the sensor's response to total darkness, (absolute black)

Figure 8.



Note that all the White through Gray to Black curves are fairly flat or straight across the visible portion of the spectrum with peaks occuring in the non visible infrared and near to ultraviolet ranges..

Figure 9 shows the data normalized so that each curve is normalized so that its maximum has a value of 1.0.

Figure 9.



Starting with Figure 10, the curves for various colored paints are presented similarly.

Figure 10 shows the raw sensor response for the various paints.

Figure 10.



Figure 11 shows the colored paints normalized to "SFE-Dolar Tree Marble White".

Note that some of the colors of paint are pure, single peak colors, whereas others show mixed response. NEP-Mayan Red, NEP-Emerald Green, NEP-French Torquoise, and Ultramarine Blue are examples of single peak colors.

Figure 11.



Figure 12 shows the colored paints data normalized so that each curve has a maxumum value has a value of 1.0

Figure 12.



And finally, Figure 13 shows a selected sensor image (Aluminum Foil), which was column summned to produce the associated raw curve, (Figures 7 and 10).

Figure 13.



At this point, I have two observations. They concern white and black pigments.

1. "SFE-Dolar Tree Marble White" seems to be much whiter than anything else. This is not surprising to me as the selection of white marble chips was the result of my observations during the construction of a Buddhist Temple near Pittsburgh. The temple is located along the Parkway East, along my way to work. This enabled me to follow its exterior construction from start to finish. When they painted the temple, its whiteness was quite striking. I inquired with an Indian colleague, and he told me that the paint was made from white marble. I remembered that. So when I was looking for white pigments, my wife and I were at our local Dollar Tree and I saw small bags pof white marble chips. The white temple came to mind. This white is the result of that.,

2. "SFE-Sychamore Black" seems to be much blacker than anything else. It is prepared from sychamore tree twigs, kilned in a homemade charcoal kiln.

All SFE pigments are sieved through a 45-micron stainless steel laboratory sieve. If you are not a "Don't use black or white" purist, you may wish to look into paints made from these two pigments. They are both natural and sustainable and provide extreme intensities of pigmentation. Just a suggestion, for what it's worth. If you have recommendations for standardizing my paint chips, I would welcome them. At that point I will rerun all my tests on the better standard chips. Hope you find this interesting, and of use. If you would like to see the results for your own or purchased paint, send me a 2.5" X 1.5" piece of Arches Aqurael Cold Pressed Paper, painted with the paint you are interested in and a second unpainted piece. PM me to get my address.