Home Creators Posts Import Register Favorites Logout

Content

This video was meant to be posted on the 4th of July, but it wasn't ready. So I'd rather take the extra time and make it right instead of just meeting a deadline.

I would love to get your thoughts on this - particularly from chemists who can weigh in on the changing color of the copper chloride flame test. Please let me know anything else you like/don't like, anything I can clarify. I really appreciate it!

Best,

Derek

Files

How Fireworks Work

Sneak preview, please post comments on Patreon.

Comments

Adrian Byram

as always great fun !! Thank you

Paul Looyen

Great video Derek. No issues I see. Entertaining and engaging as always.

Andrew

I believe the copper color thing is this: Copper hydroxide, rather than copper oxide burns blue. The copper oxide salt powder + water will make copper hydroxide but it could also be the heat, impurities, and/or a combination of these. Edit: the burning of a watery thing seems counterintuitive and I'm not the chemist you are really looking for - just brain dumping ideas :)

Hermien Pellissier

A super fun video! I do think the volume of the voice of the pyrotechnics expert is a bit lower than your voice at times though.

Veritasium

Good place to brain storm - the salt was Copper chloride and it was dissolved in methanol

Veritasium

Thanks Paul! Any chemists you could float the copper color thing past?

Tristan Ursell

Nice, great job. I was surprised to see no mention of metal chunks ... I thought many fireworks use bits of iron, magnesium, ferric oxides, etc. for particular colors / effects. I guess that was maybe covered in the discussion of the coatings.

Tommie Marshall

Amazing video man! I wrote these feedback points as I watched. Demographic context: I'm not a scientist, but I did science and math at A-Level (sort-of like APs), but I learn about science and math for fun. 1. The title is slightly awkward with "fireworks work". I know coming up with titles is a big part of acing the YouTube game, so I'm sure you've thought carefully about it, but I just wanted to note that. 2. When you show the electron micrographs of charcoal at ~3:50, you compare its molecular structure to graphite. It would be cool if you put it side by side with a micrograph of graphite to make the difference more tangible: https://www.nisenet.org/catalog/scientific-image-graphite-flakes 3. The close up gunpowder shots are awesome (~5:00), as is the drone footage at the end. 4. It could be cool to have a bit more explanation vis-a-vis the quantum mechanical aspect at ~22:30, to more clearly connect it to basic science concepts about energy and electromagnetic waves. As I understand you, the combustion raises the electrons to a higher energy level, and that because the available energy states are discrete rather than continuous, it can only 'jump' back down to its original energy level in one step, releasing a burst of energy. My questions are: a. I presume that the reason it jumps back down rather than staying there is because the higher energy level is unstable? Is that right? b. Also, the colour of the light emitted by the jump back down to a lower discrete energy level is determined by the size of the jump, which in turn would impact the frequency of radiation emitted. You mention that a particular amount of energy corresponds to a particular colour, but a brief inclusion that energy level impacts frequency which impacts colour might be useful. And I'm guessing a bigger jump would mean more energy is emitted, which means the light would be higher frequency? 5. At ~22:50 you relate the combustion of copper chloride to the quantum mechanical aspect. This helps make it tangible, but I was a little confused as to the concept of electron energy states in a copper ion vs. how that works in a compound like copper chloride. As I understand it, the Cu 2+ "donates" its two valence electrons, and two chlorine atoms get one of those each. My question is: which electrons do the jumping? I don't have a great intuitive of how ionic bonds work. Maybe this is too complex to include, but it was a question I had. If there's a way to include an explanation of that, it could close the loops between your quantum mechanics piece and the chemistry piece, which would be pretty awesome. As a side note, I can see people having questions about what Copper (II) is, and it could be a nice way to explain that. 6. In your little closing voiceover at ~29:20, it might be nice to give a nod to the historical and cultural aspect to tie those component to the science component. They're one of those rare cultural practices that seem to have proliferated in an incredibly wide array of contexts around the world, from Western New Year, to Chinese New Year, to Guy Fawkes night, to presidential transitions. Made me think of one of the opening scenes of Lord of the Rings (Fellowship of the Ring) where Gandalf sets off those magic fireworks. The link between science, history, and culture is really cool, and Veritasium does a great job of tying those aspects together. 7. The beard looks great. If I had to pick one to focus on, strengthening the explanatory link between the quantum mechanics and the chemistry of fireworks (points 4 & 5) would help people make those conceptual "ah-ha!" connections, which is the best part of learning. My favourite Veritasium videos do and amazing job of this.

Bartosz Błaszkiewicz

I am at 22:15 and so far the only thing that stands out is that blade of grass in front of cameramans leg. It looks like my screen is cracked.

John Cox

I used to be a chemistry student, but unfortunately most of the research into atomic emissions spectroscopy is gated behind journal paywalls that I don't have access to anymore. Highly annoying. Still, I was able to dig up a few free sources about the emission spectra of copper and chlorine gas. Cu: - Paper: "Emission characteristics of copper using laser induced breakdown spectroscopy at low pressure" (https://iopscience.iop.org/article/10.1088/1742-6596/1025/1/012003/pdf). On page 3 (page 4 of pdf), there is a graph showing that Cu II (that is, Cu2+, the ion in copper chloride CuCl2) has a weak but noticeable peak around 400nm. That's a bit below the the 450-495nm range that would be considered true blue, but our brain would probably consider that close enough to blue. There are stronger peaks for Cu I (that is (Cu+) around 510-515nm, which is the lower end of what we would consider green. - If you're willing to do some legwork with graphing (while the data is assumed good, the website itself is a bit primitive and doesn't show graphs, so the user has to do that themselves), this set of emission peaks from NIST might be useful. "Strong Lines of Copper": https://physics.nist.gov/PhysRefData/Handbook/Tables/coppertable2.htm Note: Wavelengths in angstroms, not nm. Cl: - A bit more difficult to dig up, but I found a free published paper from 1994: "Atomic Emission of Chlorine: Spectra from 200 to 900 nm by Sealed Inductively Coupled Plasma-Atomic Emission Spectroscopy" (https://www.researchgate.net/publication/237901081_Atomic_Emission_of_Chlorine_Spectra_from_200_to_900_nm_by_Sealed_Inductively_Coupled_Plasma-Atomic_Emission_Spectroscopy). On page 383 (page 2 of pdf) the visible emission spectra of chlorine gas is broken up into 5 graphs. The author probably did this for clarity, because the emission peaks are all over the place. The strongest lines appear at 725nm and 837nm, which is deep red. Based on this paper alone, if chlorine contributes to the flame color, it will probably be in the red. - Again, if you're willing to do so graphing, NIST provides this set of "Strong Lines of Chlorine": https://physics.nist.gov/PhysRefData/Handbook/Tables/chlorinetable2.htm Flame color: I'm going to make a semi-educated guess based off what I've dug up using free resources alone: Chlorine contributes a red (or at least dark yellow) flame. The initial intensity of the flame might be enough to cause the chlorine to lose an electron briefly to Cu2+, forming Cu+ in the plasma, which would emit light around 510-515nm (green drifting into blue). As the flame cools, Cu+ would give up that electron to plasma, oxidizing back to the more stable Cu2+, which has an emission spectra around 400nm (deep blue bordering violet). That's my guess based on non-paywalled resources.

Joel Menken

I think it’s going well. Obviously the two placeholders but I like it as it is and can’t wait to share it with my friends.

Veritasium

thanks so much for this detailed feedback!

Veritasium

yeah we did burn some magnesium but never got the right sorts of shots to fully illustrate it

Andrew

Rewriting this to address the copper chloride part rather than assume this first part; When copper chloride is heated you get copper oxide and chlorine gas. So, copper methoxide then. And depending on the concentration and amount of time given in solution the transition between green and blue may have been a combination of excess reactants burning off before getting to the higher concentration of copper methoxide and/or the heat acting, not specifically as a catalyst, but to speed up the reaction rate to create more copper methoxide and turn the flame blue.

Paul Peijzel

Hi Derek. Former assistant professor of analytical chemistry here :-) Very entertaining video, covering many subjects. It’s amazing to see what it’s like to fly through fireworks. Some comments from a first viewing: At 3:30, to be more correct, you need a source of oxygen. There is a very nice demonstration of ‘burning’ Mg in a cavity of dry-ice, reducing carbon dioxide to carbon. The reaction equation shown at 7 minutes should not be taken too seriously. Depending on temperature and amount of water in the atmosphere, the reaction products could be way different. In addition, in chemical equations, compounds should not occur on both sides of the arrow, like C and S in this case. The whole balance between the compounds taking part in the reaction(s) even depend on the geometry of the place they burn in. Evolution of gas will blow away solids in unpredictable ways. You sort of say that in the next part, where one needs to confine everything to create an explosion. Having done a lot of atomic absorption spectroscopy, I would like to bring forward that the processes in the flame are not so straightforward, and depend on temperature, oxidsing/reducing character, and even are different for certain parts of the flame. You can also see that in your experiment where the flame is partially blue, partially green. Cu forms different compounds having blue (CuCl) and green (CuOH) emissions (amongst others), there is even red CuO emission. Do we see a glimpse of that at 23:47?. Emission from the ‘desired’ compound can also be overshadowed by other emissions from the flame itself, as well as emission of other compounds also present in the flame. And to complicate stuff, if there are some solids in the flame, they may also emit thermal radiation. A candle flame turns yellow by the soot particles in it, as an example. As blue fireworks are the most difficult colour to make, you can also understand why we see this colour the least. As a general comment: be careful when mixing pyrophoric mixtures. The ones containing powerful oxidisers like chlorates, might ignite by the friction of the stirring.

Andrew

Posting this as a new comment in addition to a reply above to make sure it's seen - not for attention but to ensure accuracy for everyone (I want to know if I'm wrong as well!). When copper chloride is heated you get copper oxide and chlorine gas. Chlorine gas doesn't emit much visible light alone. But... Copper methoxide (as you stated it was in a solution of methanol which, when heated, results in copper methoxide does. And depending on the concentration and amount of time given in solution the transition between green and blue may have been a combination of excess reactants burning off before getting to the higher concentration of copper methoxide and/or the heat acting, not specifically as a catalyst, but to speed up the reaction rate to create more copper methoxide and turn the flame blue.

Balkrishna Heroor

Loved it. I have never thought about fireworks in such depth.

Rob Christian

I don't understand why we don't get to see inside the finale. You should consider doing a split screen replay so we can see inside and from the ground at the same time.