Dyes in Breakfast Cereals
While eating breakfast with my kid, they asked me about the dyes in the cereal. That got me thinking about the different dyes in food and how these dyes were analyzed. I thought that a UV-Vis instrument would be the best choice. But, what about an HPLC with a UV detector? I figured a UV detector would be able to see blue dye, but what about red or yellow dyes? Would these dyes be visible in UV light, would the signal be strong enough to see a discrete peak? There was only one way to find out.
I decided to use the cereal that we had been eating that morning, it is a puffed rice cereal with prehistoric cartoon mascots. I began with separating out the different cereals by color, a task that took quite a bit longer than I had anticipated. I figured that I needed about a 100 mL beaker full of cereal to get a good amount of dye for testing.
The Cereal being Sorted by Color
Once I had enough of each color, I went on to do the dye extraction. I crushed up the cereal just a bit and added it to a 150 mL Erlenmeyer flask, to the flask I added 20 mL of HPLC grade water and 40 mL of an ammonium hydroxide in methanol mixture (10% v/v). The flask was then placed on a hot plate and boiled for approximately 10 minutes. The solution was then allowed to cooled and the liquid was transferred to a centrifuge tube.
Blue Cereal Dye After One Extraction
The extraction process was repeated two more times to remove as much dye as possible. After the third extraction the cereal dye was all but completely extracted and the cereal had the appearance of mushy slightly colored rice. Since I was doing a qualitative identification and not quantitation of the amount of dye, I decided to stop here.
Blue Cereal After Three Extractions
The dye solution was then allowed to evaporate until approximately 25 mL remained and then centrifuged for 5 minutes at 5000 rpm. A sample of dye was then transferred to a sample vial and injected into the Lucidity LC-UV using the following method.
| Lucidity LC-UV Method | |
|---|---|
| Flow Rate | 2.0 mL/min |
| Column | C18 150 x 4.6 mm, 5μm |
| Loop Size | 5 μL |
| Oven Temperature | 35 ℃ |
| Wavelength | 260 nm ±1 |
| Flow | Gradient |
| Mobile Phase A | 32 mM Ammonium Acetate |
| Mobile Phase B | 4 : 1 Acetonitrile : Methanol |
| Gradient Program | Time (min) % B
0.0 9 1.0 9 3.5 25 6.0 33 8.5 70 8.6 9 10.0 9 |
I was looking to see if there was just a difference between the chromatograms to determine if the dye was visible. Each of the chromatograms did have a peak that was different than that of the other colors, although there were small underling peaks in each of the chromatograms. I think that because I couldn’t get a perfect separation of the rice flakes, some colors were stuck together, and each flake rubs against one another in the box, that some of the dye was transferred to each flake. So, there was a bit of red and yellow in the blue and so forth.
Chromatogram of the Blue Dyed Cereal
Chromatogram of the Red Dyed Cereal
Chromatogram of the Yellow Dyed Cereal
Overlaid Chromatogram Showing the Unique Peaks for the Dyes
Looking at the overlaid chromatogram I was able to see that there were unique peaks that only showed up for that color cereal. These peaks are shown with the colored arrows, and the colors of the arrows correspond to the dye color. The red and yellow dye peaks were much larger than the blue dye peak. This may be due to blue dye being a stronger pigment. I know from my own experience that to color something with blue food dye, a little goes a long way, where as red needs more to not look pink, and a whole lot of yellow is needed for it to even show.
While these chromatograms may have shown that the dyes were present, I couldn’t be 100% sure there wasn’t something else showing in each of these chromatograms. So, I went back to the store and bought food dye that was the same as that in the cereal, red #40, blue #1 and yellow #5. About 0.2 g of the dye was added to a sample vial and 1.5 mL of water was added. The vials were shaken for approximately 1 minute to mix the dye and water. These dyes were injected into the Lucidity LC-UV using the same method.
Overlay of the Chromatograms of the Food Dyes
Where there were differences between the cereal colors, the pure dye colors showed the same differences, just much more pronounced. This led me to believe that the differences in the cereal chromatograms is from the dye and the dye alone.
As a follow up, the cereal had not only blue, red and yellow colored pieces, but orange, purple and green. Green dye is an isolated color but the ingredient list doesn’t show any green dye being used, so I am assuming that all of the secondary colors are due to color mixing of the dyes. I plan to extract these dyes using the above procedure and see if there are two hits for the different dyes in the resulting chromatograms.