HomeНаука и техникаRelated VideosMore From: Tommy Technetium

Smartphone "Spectrophotometer"

187 ratings | 36801 views
This is a simple experimental set up that allows students to visualize the processes involved in spectrophotometric measurements. As a bonus, excellent results are obtained. Simple materials and a smartphone are used to measure concentrations of Red Dye #40 in Strawberry PowerAde Zero. The experimental design and data analysis mimics several basic features of involved in absorption spectrophotometry and Beer's Law analysis.
Html code for embedding videos on your blog
Text Comments (80)
ZodaChem (4 months ago)
This is more a smartphone colorimeter.
mayorc (5 months ago)
Any way of using similar approach to sample Reflectance (a series of reflectance values at various wavelenghts) Curve of a color sample on paper to mimic an expensive spectrometer used for measuring color and pigments reflectance ? Would make a great and inexpensive tool.
Tommy Technetium (5 months ago)
That's a good idea. I'll have to think about this for awhile. My hunch is that a similar approach would not work for reflectance, but of course I haven't tried...
Don Dan (5 months ago)
That lock is the Exposure Lock, what it means is just the phone won't change shutter speed or iso while taking the measure (usually the camera does that to adapt to new light conditions).
Don Dan (5 months ago)
Glad I could help!
Tommy Technetium (5 months ago)
Thank you, Dan! This is extremely helpful.
Soan (7 months ago)
did not find the coloremeter app in google store.....suggest other which is available
Tommy Technetium (7 months ago)
We have used this one with good success: https://itunes.apple.com/us/app/color-analyzer-get-info-from-image-with-camera/id1160206848?mt=8
Thiago Ribeiro (8 months ago)
Interest video. But I have a comment about your data analysis: we can notice that a higher concentration results in a lower intensity. If you measured that an intensity of 111 would give you 4.83 of concentration, a higher intensity of 116 have to give you a lower concentration than 4.83. However, from your data analysis you have found it to be slightly higher (4.85). This is due to the linearization you have done, converting into absorbance and extracting a straight line. In this case, there is actually no need for converting into absorbance at all. Using a 4th grade polynomial equation in Excel to correlate between intensity and concentration resulted in a R=1 equation that was used to calculate the concentration of 3.74 for an intensity of 116.
Tommy Technetium (8 months ago)
I used the known molar absorptivity of Red Dye #40 and Blue Dye #1 found in the reference listed here:(http://pubs.acs.org/doi/abs/10.1021/ed100545v). I then used our Hitachi U-2900 absorption spectrometer to measure the absorbance of the Fruit Punch PowerAde Zero and Blueberry Pommegranate Gatorade. The concentrations of Red Dye #40 in the PowerAde and Blue Dye #1 found in the Gatorade were then found using A = ebc (A is absorbance, e is molar absorptitivity, b is pathlength of cuvette, c is concentration). The results for the concentrations I found using this method were very close to what was found at this article: (Amounts of Artificial Food Colors in Commonly Consumed Beverages and Potential Behavioral Implications for Consumption in Children: Revisited September 2013 Clinical Pediatrics 53(2)). Again, thank you for your questions and insight and let me know how this works out in your home brewing!
Thiago Ribeiro (8 months ago)
I totally agree with you that the general behavior of the physical-chemistry of the 4th order polynomial would not be representative at all (unless infinite points are given), even with the best fit possible. But I often find that for a small data analysis, using Beer-lambert law its just unnecessary and could lead to inaccurate results, but could be extrapolated out of the data range (of course, taking in consideration possible non-linearities in higher concentrations) because it does match the physics, while the 4th order polynomial doesn't. Anyway, great stuff and I will definitely be using your method on my home brewing system for color analysis. Do you mind sharing how you measured the first concentration of the Power Aid?
Tommy Technetium (8 months ago)
Hi Thiago, thank you for your question; great observation! You are indeed correct that a higher concentration results in a lower absorbance; this is observed quite often in experiments such as the one presented in the video. You should note, however, that a 4th order polynomial will perfectly fit any set of 5 data points (try it an see). So while fitting the 5 data points with a 4th order polynomial gives me an excellent fit, it does not give me much confidence that a 4th order polynomial faithfully represents my data. In the data analysis displayed in the video, I am using Beer's Law, which is often used by chemists to fit absorbance vs. concentration curves. Occasionally we will use a second order polynomial to fit the data, because as you note the data tend to become non-linear at higher concentrations. If we wished to fit data to a 4th order polynomial, we would need to take a very large number of data points to gain confidence in our fit. Generally this is not done.
Thiago Ribeiro (8 months ago)
I am curious about how you measured the concentration of the dye in the first standard of power aid, before the dilution. Could you share that info with us?
Natasha Milena (9 months ago)
Can you give me your email address? I would like to ask to you somenthings
Tommy Technetium (9 months ago)
You should be able to find my email at the following site: https://www.arbor.edu/faculty/tom-kuntzleman/
Kee Dee (10 months ago)
great !
Reb Perez (1 year ago)
sir? Good day :-) 1. If my sample is KMnO4 (purple in color) what should I use as a background color? 2.What if my sample is blue in color then? 3. can I use the white color instead as a background for all colors?
Reb Perez (1 year ago)
Thank you sir for responsed 😁 We plan to build a set-up for this Idea.Turning it to instrument type. It is subject for our experiment as an alternative way to measure absorbance. Having different color solution, we expect to have a different background also. What can I use instead for conveniency, so that I will not change the background screen everytime I change the color of sample solution?
Tommy Technetium (1 year ago)
Hi Reb, good question. If you look up the absorption spectrum of KMnO4, you will notice it absorbs very well in the green region. Therefore, I would use green as my background color if analyzing KMnO4. I have achieved very good results using a red background for blue colored solutions. Let me know how your experiments work out!
→ to the knee (1 year ago)
This is a pretty nice explanation of calibration in general. Thanks for publishing this video!
Tommy Technetium (1 year ago)
Thank you for watching and commenting. I also appreciate the feedback.
Firefly Sci (1 year ago)
Hi Tom. Wow this is really cool! I love the cuvette box, have you tried it with using an actual cuvette? I'm wondering if the data would be more accurate if you used a flat surface cuvette vs the round cup.
Tommy Technetium (1 year ago)
Thank you, Firefly Sci. We have used actual cuvettes with good success. We have also used small test tubes. I have not done an in-depth analysis of whether the results are more accurate with one vs. the other. I imagine cuvettes would work the best, but as you can see we still get remarkably good results using just cups. If you like, you can learn more in our article published here: http://pubs.acs.org/doi/abs/10.1021/acs.jchemed.5b00844
남유진 (1 year ago)
Hi! Thanks for your nice video. I would like to ask you why the background color should be complementary color of the solution!
Tommy Technetium (1 year ago)
2. First you will need to make sure you know which particular compound in the extracts you are looking for. Then you can proceed two ways: a. find the molar absorptivity of those compounds, and use Absorbance = (molar absorptivity) x (concentration) x (pathlength). Or, b) buy the pure, known compounds and make up solutions of these compounds at known molar concentrations (called standards). Then construct a Beer's Law plot of the absorbance of each standard versus concentration. Compare the absorbance of your extract to the Beer's Law plot, and you'll get the concentration of a particular component in the extract. That is assuming, however, that nothing in the extract is interfering with the compound you are looking for...
Tommy Technetium (1 year ago)
1. See the comments section at the blog post here: (https://www.chemedx.org/blog/use-your-smartphone-absorption-spectrophotometer ) More specifically, I used the known molar absorptivity of Red Dye #40 and Blue Dye #1 found in the reference listed here (http://pubs.acs.org/doi/abs/10.1021/ed100545v). I then used our Hitachi U-2900 absorption spectrometer to measure the absorbance of the Fruit Punch PowerAde Zero and Blueberry Pommegranate Gatorade. The concentrations of Red Dye #40 in the PowerAde and Blue Dye #1 found in the Gatorade were then found using A = ebc (A is absorbance, e is molar absorptitivity, b is pathlength of cuvette, c is concentration). The results for the concentrations I found using this method were similar - to within the error I cited in the blog - to what was found at the reference found here (https://www.researchgate.net/publication/256612178).
남유진 (1 year ago)
Tom Kuntzleman Thank you. Sorry but I have two more questions. 1. How did you determine the molarity of powerade? 2. And in my experiment, I am using several extracts from onions, watermelons, etc. Then how can I know the molarity of them?
Tommy Technetium (1 year ago)
1. Try this site for specific wavelengths: https://academo.org/demos/wavelength-to-colour-relationship/ 2. If the solution is colorless, then the absorbance will be zero. 3. If you see color changes during the course of the experiment, then this method will work. Otherwise, it won't.
남유진 (1 year ago)
Tom Kuntzleman Thank you for your answer! I have more questions. I'm doing an experiment about antioxidant effacement of several extracts from food. For this experiment, I heard that I need to measure the absorbance of each solution. 1. In previous researches, they said I need to measure the absorbace level at specific wavelength(like at 510nm). So I wonder how I can regulate the wavelength with this spectrophotometer. As complementary colors for each solution are different, I think the wavelength would be different as well. Is there any advice for my situation? 2. How can I measure the absorbance if the solution is clear? For some extracts, they have no color. 3. I'm not sure if the results (absorbance measured by this) can exactly show the antioxidant effacement. What's your opinion about this? Thanks for reading these long questions:-)
FASB CPA (1 year ago)
How can I use your analysis in a practical situation? Is red dye #40 bad for humans?
Jackelyne MV (10 months ago)
How did you get the concentrations?
Tommy Technetium (1 year ago)
Hi FASB CPA. This analysis could be used to detect just about any colored compound, as long as you could make up known concentrations of the compound of interest. It is safe to consume the amount of red dye 40 that is placed in foods. The LD50 for Red 40 is greater than 10 grams per kilogram of body weight. So an average person would need to consume enormous amounts of red 40 (roughly a kilogram) in one sitting to experience ill effects. You can learn more by looking at the msds for red 40 here: http://www.sciencelab.com/msds.php?msdsId=9924020
valor36az (1 year ago)
Thank you for taking the time to post this excellent example
Tommy Technetium (1 year ago)
Thank you for commenting, valor36az.
Adrian Serapio (1 year ago)
Sir Tom Kuntzleman! Hey there! I am interested of using this for this application. Please do comment for your suggestions on feasibility, among other things to point out. I am going to utilize this mechanism to create a smartphone app capable of detecting body fat level. Using the smartphone camera light, the light, after passing through a pinched part of the arm, (causing a reddish glow), will be detected using the colorimeter. The concentration will be based on the caliper reading. Using the formula for absorbance, this will be plotted into a graph. To test this, there will be samples without using the caliper, and simply relying on the formulas to find out the percentage level of body fat. :) Please answer ASAP!!! 😄
Tommy Technetium (1 year ago)
Hi Kenneth. Using a smart phone to detect % body fat is an interesting idea. However, at this point I can't imagine how to do this using the method described in this video.
Adrian Serapio (1 year ago)
+Tom Kuntzleman
Maricela Nava (1 year ago)
es muy es tupido
LordBelial 676 (9 months ago)
Maricela Nava como tu analfabetismo :v
9B 2020 (1 year ago)
Hello Sir Tom! I am currently incoming grade 10, and would want to explore the possibilities of measuring hemoglobin concentration using spectrometry. I have tried researching the Beer's Law and how smartphones using IR sensors were already tried for and were successful in measuring hemoglobin levels. These are noninvasive techniques that explore wavelengths at 650nm, and of which the amounts absorbed, and through machine algorithms measure the hemoglobin content. I want to investigate this thing by recreating a similar test, this time, in hopes of maybe even measuring other components such as maybe glucose, or maybe even cholesterol? However, I'm not really sure where to start. I saw your video and am humbly asking if this method may be applicable to rudimentarily measure hemoglobin percentages as compared to known hemoglobin percentages, and graph it? What do you think sir? :)
9B 2020 (1 year ago)
Thank you :)
Tommy Technetium (1 year ago)
Hemoglobin appears red in solution, which means it will absorb green light well. Therefore I think that you could, in principle, use this technique to measure hemoglobin concentrations. The trick would be you would have to purchase pure hemoglobin to make your standard curve - and I imagine it would be quite expensive to do so. Also, the technique you see in the video is not likely to be useful in a non-invasive technique.
RIJA ZAFAR (1 year ago)
You mention "the color of sheet must be closely matches the complimentary color of solution" why is that? and secondly what exactly the value of R, G, B represents in that app??
RIJA ZAFAR (1 year ago)
thank you!
Tommy Technetium (1 year ago)
1. The light reflected off of the sheet should be easily absorbed by the solution. This is best accomplished when the color of the sheet is complimentary to the color of the solution. 2. The R, G, and B values represent the number of R, G, and B pixels required to match the color of the pixels inside the circle displayed on the app screen. The color inside the circle represents an average color.
Sandy Ramanjooloo (1 year ago)
Thank you for this experimental set up. Great I would like to know how to measure Deuterium concentration in ppm in water. Thank you
Sandy Ramanjooloo (1 year ago)
Thank you
Tommy Technetium (1 year ago)
Sandy, try looking through the papers mentioned here: https://www.researchgate.net/post/How_to_measure_deuterium_in_water_qualitative_and_quantitative
Sandy Ramanjooloo (1 year ago)
Thank you Tom, but do you know which instrument or machine can detect deuterium at his best
Tommy Technetium (1 year ago)
Thank you for your comments, Sandy. I can't think of a way to colorimetrically detect deuterium in water, so I doubt this method will work for that.
I need to know what app you used to perform the analysis on the smart phone. Thank you
Thank you so much
Tommy Technetium (1 year ago)
On the iphone we use "colorometer". On an Android, try "ColorMeter Free".
Akash Manoj (1 year ago)
Hello! Thanks for your helpful video. Can I quantify the absorbance of proteins in the biuret solution using this smartphone based spectrophotometer? Are the results accurate in this technique?
Tommy Technetium (1 year ago)
I imagine you could do this. What color develops when the protein interacts with the biuret solution? Yes, we find one can get quite accurate results when using this technique along with proper sample set up.
Emelia B (1 year ago)
Could you just use different colors for the water instead of drinks?
Tommy Technetium (1 year ago)
Yes, I have done this experiment using food dyes at different concentration. When using various colored dyes, be sure to use the correct complimentary color for the background.
Mukund Seshadri (1 year ago)
Hi, Mr. Kuntzleman. I am in the ninth grade and using this app for a research project. I tested a multitude of solutions of water containing different levels of lead in them, and my initial data showed that as the level of lead went up, each RGB value increased as well. Since my solution is clear (water w/ lead), I used the colorometer to test each solution 3 different times - once w/ each primary color background as there is technically no complimentary color to clearness. With every background color- the corresponding color value increased as the level of lead increased, so I was very encouraged. However, when I went to test the data a second time - to ensure accuracy, the R, G, B values were quite varied in comparison to my initial data. I feel that this is a problem as I would like to use this to present as a household way to test for lead in water. Any ideas as to why there wasn't consistency with the readings? Would you be able to help? Thanks!
Mukund Seshadri (1 year ago)
Thanks so much for the response - i sent you an email.
Tommy Technetium (1 year ago)
Lots of good stuff here, Mukund. I think you will need to somehow react your samples with something to get a color change. Colorless solutions won't work with the cell phone spectrophotometer. One possibility for you to get color is to react the lead samples with sodium rhodizonate. See http://pubs.acs.org/doi/abs/10.1021/i560110a034?journalCode=iecac0. However, reaction of lead with rhodizonate might form precipitates, which might throw off your results. You might try adjusting pH to avoid formation of precipitates. Now for your questions. Perception of color is a tricky phenomenon. What you have said about us observing light that is reflected is correct. Keep in mind, however, that if an object reflects blue and red light simultaneously, it will likely be observed as some shade of purple (red + blue = purple). With solutions, we observe the light that passes through the solution (is transmitted through the solution), but not the light that is absorbed by the solution. To a good approximation, the color a solution absorbs will be complementary to the color it appears. Thus, a red solution absorbs green light very well, and a green solution absorbs red light very well. Yellow solutions absorb violet light (and violet solutions yellow); blue solutions absorb orange light (and orange solutions blue). These are not hard and fast rules, but rather guiding principles. In a spectrophotometer, light from a source is sent through samples and then detected. We mimic this in the cell phone experiment. In the cell phone experiment experiment, light that is reflected off of the background paper is sent through the samples and is detected by the cell phone RGB analyzer. If your samples have no color, then essentially no light will be absorbed. If you treat your lead samples with sodium rhodizonate, it is likely you will get a red colored solution. If this is the case, use a green background, and measure the G value in the RGB analyzer. This way, you'll be measuring the green light that is transmitted. If you google my name you should be able to find my email at the university where I work. If you email me, I will be happy to send you and your teacher a lot more information on this.
Mukund Seshadri (1 year ago)
Thank you so much for the response! Sorry in advance as well - this response might be a little long! I created my lead solutions with my AP chemistry teacher. He had a lead compound I was able to use for the solutions which was lead nitrate. I calculated the molar mass of lead nitrate(331.2098) and then the molar mass of lead(207.2). I then divided the molar mass of lead nitrate by the molar mass of lead(331.2098/207.2) and got the number 1.5985. I made solutions at 15,30,45,60, and 75 ppm so I then multiplied 1.5985 by 0.015, 0.03, 0.045, 0.06, and 0.075 to obtain the amount of lead nitrate needed to be dissolved in 1 Liter of water to get that amount of lead. For ex : 0.015 * 1.5985 = about 0.024 or 24 ppm - therefore, 24 ppm of lead nitrate needs to be dissolved in 1 Liter of water to obtain 15 ppm of lead in that solution. I then just measure out those values and made the solutions with 1 L of distilled water. As for the colorimetric science behind this, my thought process is that as the lead increases in the solution being tested, it would make sense that the RGB values increase, as lead is a heavy metal and would increase the absorbance level of the light. I do not think I really see any slight color in my solutions - it looks pretty clear to me. Although, I did read an article talking about how scientists do believe that the closest tint/color to water is a very slight/light blue. I do have two quick questions as well - 1. Is this colorimeter measuring light that is being reflected or absorbed? - I learned in chemistry that the light we see, is light being reflected, and every other color is being absorbed by that object so we don't see any color except for what is being reflected. Is that correct? 2. Is the background paper that is complimentary to the color of the solution being used to control absorption or reflection of light?
Tommy Technetium (1 year ago)
Hi Mukund. I am quite impressed with your work. I'm actually a little surprised that you were able to detect increases in lead if your solutions were colorless (We chemists like to say that, clear is not a color - red Kool-Aid is clear. Clear means transparent, and transparent doesn't necessarily mean no color). My hunch is that something else was affecting your RGB values other than lead. If your lead solutions contained no color, then I don't see how RGB values could reliably with lead concentration. Could you see ANY color in these lead solutions at all? I am specifically interested if perhaps you observed a very faint blue? Could you send me a little more information regarding how you prepared your samples that contained lead?
diana dwi hapsari (1 year ago)
can you explain the all equations comes from? and give me a references? thx very much
Tommy Technetium (1 year ago)
Great; glad to help
Bimo Pratama (1 year ago)
thank you so much tom last week i use my sister id to ask the question to you diana is my sister
Tommy Technetium (1 year ago)
Absorbance =-log(I/I0) comes from the definition of absorbance. Absorbance is simply a measure of how much light DOES NOT get through the sample. It turns out that absorbance is linearly related to concentration, thus absorbance = m(conc) + b (think y = mx + b from algebra). You can read about this more in depth in Quantitative Chemical Analysis, 7th edition by Daniel C. Harris (p. 382). You can actually get a free pdf copy of this online here: http://www2.fc.unesp.br/lvq/LVQ_experimentos/analitica_qualitativa/material%203.pdf me know if you have questions.
Bimo Pratama (1 year ago)
what is the application on your iphone? thx
Tommy Technetium (1 year ago)
It is called "colorometer".
Tyler Voyer (1 year ago)
I know this is an old video but I am doing a project based on this video and is there a way you can tell me how you found the molarity of the Powerade? I understand what it is and how you would calculate it IF you had the moles of the Powerade. I know you have the volume but how do you know how many moles you have? Thanks Tom, great video!!
Tommy Technetium (1 year ago)
By diluting known volumes of the Powerade with known volumes of water. For example, if 1 mL of powerade was diluted to 10 mL total with water, the resulting solution was 0.0000095 M (9.5 micro molar).
Tyler Voyer (1 year ago)
Tom Kuntzleman and you found out the molarity when you diluted it how?
Tommy Technetium (1 year ago)
+Tyler Voyer That is right, the molarity of Red Dye #40 in Fruit Punch Powerade Zero is about 0.000095 M.
Tyler Voyer (1 year ago)
So basically what you're saying the website is tha byou know the molarity of the Fruit Punch Powerade Zero is 95 +or- 5 x 10^-6 M? If so two questions. One, how did you know this? Two, how did you figure out the molarity once you diluted the solution? Did you divide the "95+or-..." by the solute (water). Thanks AGAIN!
Tommy Technetium (1 year ago)
Hi Tyler, thank you for your question. You can see the Powerade concentrations listed in If "Further Tips #5" found in the post at this link: https://www.chemedx.org/blog/use-your-smartphone-absorption-spectrophotometer. If you scroll down to the comments, you can see how I determined the various Powerade concentrations. Let me know if you need further clarification. Let me know how things work out in your project!
Alvin Moore (2 years ago)
Tom, great video! I liked the way you walked us through the data analysis as well, rather than just assuming anyone who cared enough to do the data analysis could probably figure it out. I am puzzled by one thing in the data. Your unknown comes out with a higher G value than the nearest point on the std curve, which would make me think it's conc should be a little less. However, your calc conc of the unknown comes out a little higher than that of that nearest point. Am I missing something obvious here?
Tommy Technetium (2 years ago)
Hi Alvin, great question and thank you for commenting. You are right. Comparing the G value of the unknown (116) with the nearest G value on the curve (111) leads one to believe the concentration of the unknown should be a little less than 4.83 micromolar. However, the comparison of the unknown is done with ALL of the points on the curve. All of the points on the curve yield the equation y = 0.0592x + .0093, and this equation is what was used to find the unknown concentration.
Francine Morris (2 years ago)
Ummm....I hope this doesn't sound inappropriate but I LOVE YOU! I am a postdoc and I have recently started teaching undergraduates part-time as well as volunteering for science outreach programs that work with h.s and middle schoolers. This is awesome! Thanks so much for sharing:) I guess the only downside is restricted to read samples one at a time rather than all at once.
Impact Hyper (2 years ago)
+Tom Kuntzleman pls stop whit youtube you so bad!!!!!!!!!
Tommy Technetium (2 years ago)
+Francine Morris Thank you for your kind words, Francine. I'm glad you have entered the world of teaching and science outreach! Please do let me know if you figure out any extensions / improvements to this experiment or any others I have posted. I suppose one could measure more samples at once if one constructed a sample box appropriately. I never thought of that before; thanks for the inspiration. Best wishes, Tom

Would you like to comment?

Join YouTube for a free account, or sign in if you are already a member.