Chat with Greg Fridman

Follow-up to my last smoke and exhaust episode. Two viewer questions in one: why is bread exhaust alcohols, and can plasma clean cigarette smoke in your apartment? Yes to both, but cigarette smoke has a surprise you probably haven't thought about.

⏱️ Chapters

00:00 The two viewer questions in one

00:44 3 topics: VOCs, byproducts, and "some crap plasma can't remove"

01:15 Why bread exhaust is alcohols: yeast eats sugar and produces ethanol

01:40 Plus aldehydes, organic acids, esters (this is why fresh bread smells amazing)

02:14 One loaf smells wonderful. A factory produces clouds.

02:39 VOCs are easy for plasma to oxidize (alcohol burns readily)

02:58 Now cigarette smoke, which is the most studied smoke on earth (tobacco lawsuit funding)

03:22 Cigarette VOCs: easy. Nitrogen oxides from the thermal process: harder.

03:43 NO reacts with plasma-generated ozone to make peroxynitrite (which you also then need to manage)

03:55 The byproduct problem in both cases

04:12 What do you do with the ozone and peroxide plasma itself generates?

04:25 Cigarette smoke also has hydrogen cyanide and ammonia

04:37 Now you're putting an air scrubber in a house with kids in it

04:59 Filtration choices matter a lot

05:08 Point 3: some contaminants plasma just can't remove

05:14 Bread exhaust is mostly clean-able. Cigarette smoke is not.

05:41 The tar problem: particles/droplets are too big for non-equilibrium plasma

05:56 Plasma works one molecule at a time. Tar is thousands of molecules per particle.

06:16 One approach: use plasma to charge droplets, then capture with electrostatic precipitator

06:32 My favorite hidden problem: tobacco plants absorb radioactive isotopes from soil

06:40 Radioactive isotopes are in cigarettes, then in the smoke, then in the air

06:53 Trapping those is its own challenge

07:18 In a home system with children around, this all compounds

07:45 Commercial cigarette-air-cleaning systems use 5-6 filters in series

07:48 Plasma for VOCs, HEPA for particles, activated carbon for what plasma missed, pre-filter for dust, sometimes electrostatic precipitator on top

08:13 So: hire a plasma engineer or consult with one

08:23 Recap

08:34 VOCs from bread and cigarettes are both easy for plasma to remove

08:40 Byproduct management is the real engineering challenge

08:51 There is no magic. Plasma is a strong oxidizer. Some things it can't remove.

If you're building an air purification product, working in industrial exhaust, or you were about to buy a "plasma cigarette smoke cleaner" for your apartment, this is the one to watch first.

🔔 Subscribe for more in the "Can Plasma?" series: @gregfridman

#CanPlasma #PlasmaScience #AirPurification #IndoorAirQuality #CigaretteSmoke #IndustrialExhaust #Filtration #EnvironmentalEngineering

Next in the "Can Plasma?" series: can plasma clean a surface without touching it? Yes. Reactive gas-phase species do the work with zero physical contact. Great for organic contamination. Bad for big chunks and metals. Fast enough for Kodak to use on film production at 8 meters per second.

⏱️ Chapters

00:00 The question, and 3 things to cover

00:07 Reactive species vs physical contact

00:20 What plasma can and can't remove

00:25 Why this matters industrially

00:31 Reactive species work one molecule at a time

00:51 Big chunk of contamination? Plasma is not your tool. It'll take forever.

01:06 Where plasma shines: low-level contamination on a surface

01:12 Gas-phase reactive species near the surface, no contact required

01:30 What plasma removes well: organic contaminants

01:37 Dielectric barrier discharges are especially good at this

01:46 Oils, proteins, fingerprints

01:57 What plasma can't remove: metal salts, metals, large chunks

02:14 Any organic contaminant? Plasma is fantastic.

02:20 Industrial case study: my Kodak collaboration

02:33 Engineers from Kodak, the last generation of Kodak film (APS, Advanced Photo Systems)

02:49 That film was made on a web (continuous roll fed through processing)

02:56 Look up "web press" vs "sheet-fed press" if you're curious

03:11 Kodak ran their film through plasma at 8 meters per second

03:31 Two goals: clean the film, and make it hydrophilic so coatings stick better

03:58 Kodak managed all of it at 8 meters per second

04:02 Touchless matters when you're moving material that fast

04:05 No scraping. No physical modification. Gas-phase only.

04:15 Alternative would be spraying chemicals on it

04:22 But chemicals are sticky, and you then need to neutralize them

04:28 Reactive plasma species have sub-second lifetimes, so nothing to neutralize downstream

04:36 Recap

04:41 Reactive species vs physical contact

04:47 Organics yes, metals and chunks no

04:59 Touchless matters at high industrial speeds, and it's a repeatable process

If you work in continuous-process manufacturing (film, packaging, textiles, semiconductors) or you've wondered how you clean a surface at meters per second, this is the episode.

🔔 Subscribe for more in the "Can Plasma?" series: @gregfridman

#CanPlasma #PlasmaScience #SurfaceCleaning #WebPress #Manufacturing #SemiconductorManufacturing #Kodak #IndustrialEngineering

Next in the "Can Plasma?" series: can plasma clean smoke and exhaust? Yes, it's a proven after-treatment technology. Three catches: soot is hard, ozone is a byproduct, and electricity costs more than natural gas.

⏱️ Chapters

00:00 The question, and 3 things to cover

00:10 Non-thermal plasma is a proven exhaust after-treatment technology

00:19 Catch 1: soot and particulate matter

00:29 Catch 2: byproducts (ozone) and electricity cost

00:52 The Campbell Soup / Pepperidge Farm story

01:07 Baking thousands of loaves of bread produces alcohol vapors

01:37 Their current solution: burn off residual smoke with methane flame

01:47 The trap: burning methane uses your CO₂ credits, buys natural gas, adds compressed-gas safety overhead

02:16 Plasma becomes an attractive alternative

02:25 The soot problem: exhaust with unburned diesel or fine particulates

02:37 Non-equilibrium plasma is too slow: breaks down molecules one at a time

02:55 If it's volatile organics only, no problem. Soot changes the game.

03:07 A mentor of mine at MIT commercialized a plasma afterburner for diesel truck exhaust

03:23 The story: idling semi truck outside my student apartment all night

03:36 Loud, smoky, running the AC on the diesel

03:55 Gliding-arc (transitional) plasma is higher power and can actually burn soot

04:06 The trade-off: electricity cost

04:09 Not commercialized for diesel trucks yet because you'd need to carry a battery for the plasma

04:21 Catch 3: ozone byproduct

04:35 The FAA airplane story

04:41 We can plasma-clean airplane cabin air (enclosed environment, people sneezing/coughing)

04:59 But ozone byproduct plus plasma as ignition source with fuel leaks = no-go

05:13 Recap

05:21 Yes, it works. The 3 catches are soot, ozone byproducts, and electricity cost.

If you work in emissions control, diesel exhaust after-treatment, industrial ventilation, or aviation, this is the map episode.

🔔 Subscribe for more in the "Can Plasma?" series: @gregfridman

#CanPlasma #PlasmaScience #EmissionsControl #ExhaustAfterTreatment #Diesel #IndustrialVentilation #AirQuality #EnvironmentalEngineering

A viewer asked how I deal with tangent projects. The ideas that pop up in the middle of what you're actually supposed to be doing. I have a 3-step algorithm I use every day. Here it is.

⏱️ Chapters

00:00 The question: how do you handle tangents?

00:32 The 3-step algorithm: recognize, delegate, delete

01:05 Step 1: recognize which ideas are actually important

01:09 Every intake needs a filter (Slack, meetings, your own head)

01:21 Important for you? The team? The company? Your house?

01:37 My system: Todoist, with a folder called LTS (long-term storage)

02:04 Borrowed the name from Amazon's cloud storage tiers

02:08 Monthly ritual: first Monday or Tuesday, review the LTS list

02:21 Do it, delegate it, delete it, or postpone. But make the decision.

02:44 Anti-example: task was to paint the door, then your house burned down. Delete.

02:53 Step 2: delegate if possible

02:56 You can delegate to your future self, but be careful

02:59 The "record a video" trap: first need to buy a camera, set up mic, monitor, lighting, set up, set up, set up

03:22 Get too sucked in and you never record

03:35 Delegate to future self wisely

03:45 The honey-do list criteria (things the honey needs to do for the wifey)

03:57 Am I technically capable? Am I physically capable? Am I the best choice?

04:10 What does it cost me vs hiring someone?

04:12 Current example: finishing a patio

04:20 Absolutely within my technical capability

04:37 Would consume entire weekends and evenings of physical labor

04:52 Delegate to a paid professional. Even things you could do, sometimes shouldn't.

04:57 Colleagues also have costs (their time and company time)

05:23 Step 3: delete ruthlessly

05:32 Something felt super important at some point in time

05:40 I try to note WHY, so I can revisit the reasoning honestly later

05:46 Todoist lets you attach images and voice notes

05:54 Delete anyway

05:58 A boss I had years ago carried "super important tasks" on a crappy little piece of paper in his shirt pocket

06:10 Things like "go talk to the president"

06:26 One day he showed up gray, said "I lost the piece of paper"

06:39 Then, calmly: "Fuck it. It's important enough, they'll find me."

06:46 Great question to ask yourself before doing something you thought was urgent

06:56 Recognize, delegate, delete. That's the algorithm.

If you're a founder, PhD student, engineering lead, or anyone whose inbox is smarter than they are, this one's for you.

🔔 Subscribe for more: @gregfridman

#Productivity #Founders #TimeManagement #Todoist #Focus #Delegation #Leadership #Research

Next in the "Can Plasma?" series: can plasma make nanoparticles? Yes. And a personal story about how I accidentally discovered some in grad school, why my advisor made me drop it, and the focus lesson that stuck with me.

⏱️ Chapters

00:00 The question, and a story to go with it

00:11 The quick technical answer: yes, from many precursors

00:19 Flame and plasma are both good candidates for a reactive synthesis environment

00:31 Oxidize, atomize, condense onto surfaces to grow nanomaterials

00:59 The nano boom era: every proposal we wrote had "nano" in the title

01:06 The chemists laughed because they were working in Angstroms (sub-nanometer)

01:18 So yes, plasma can make nanoparticles

01:23 The real answer depends on which particle, what purity, what reproducibility

01:30 The story starts back in my master's days

01:33 My setup: plasma-treated water droplets, electric-field deposition

01:47 A nanoprinter for proteins and biomolecules (in argon to preserve the biology)

02:19 My chamber was built mostly with acrylic

02:21 When I screwed something up, plasma would arc onto the acrylic

02:29 It built these fine fibrous black ropes between the electrodes

02:44 At the same time I was training on SEM, TEM, XPS

03:02 So I zoomed in on the ropes

03:14 What I found: nanometer-scale triangular shapes

03:27 I got very excited

03:28 Ran to my PhD advisor (one of my seven)

03:42 "God, my system synthesizes these really cool nanoparticles!"

03:48 Advisor: "What's your department?" Bioengineering.

03:54 "What's your project?" Printing biomolecules.

04:02 "And which part of that is nanomaterial synthesis?"

04:08 None.

04:14 The lesson: a PhD is focused study. Easy to get distracted. Nanomaterials are cool. Not my field.

04:37 So I abandoned it. But I still remember those shapes.

04:43 I'll try to dig up a photo and add it here.

04:49 Bottom line: yes, plasma can make nanoparticles.

04:54 Ask me a more specific question and I'll go deeper. The field has advanced very far.

If you're a grad student, PI, or anyone who's ever gotten distracted by a shiny side result, this is the episode.

🔔 Subscribe for more in the "Can Plasma?" series: @gregfridman

#CanPlasma #PlasmaScience #Nanoparticles #Nanotechnology #PhDLife #Research #MaterialsScience #GradSchool

Next in the "Can Plasma?" series: can plasma 3D print metal? Yes. And that's just one of the eight distinct places plasma already shows up in additive manufacturing. Quick tour.

⏱️ Chapters

00:00 The question, and the broader ask

00:20 1. Feedstock production: plasma melts wire and atomizes molten metal into fine droplets

00:35 Control atmosphere to reduce oxygen content

00:41 Blend different wires into plasma for on-the-fly alloy creation

00:54 2. Feedstock conditioning: clean rough particles, spheroidize them, ready for laser sintering

01:19 3. Instead of a laser, use plasma directly: arcs or thermal plasma melting the metal in place

01:37 4. Surface activation and modification for printing (including between layers)

01:49 5. Post-processing: plasma as a heat source to re-melt or polish surface

02:02 6. In-situ diagnostics during laser sintering (the underdiscussed one)

02:10 The laser creates plasma anyway. Use optical emission spectroscopy on it to monitor the print.

02:27 7. Chamber atmosphere control: enrich reactive species, clean impurities

02:47 Same tools used in microelectronics

02:54 8. My favorite: targeted metal deposition via plasma-charged droplets

03:01 I've spent most of my career on droplets and sprays in plasma

03:10 Anything put into atmospheric plasma charges negatively (happy to make a full video on why)

03:25 Once charged, you can steer with an electric field

03:35 Molten metal droplets, plasma-charged, deposited with precision

03:42 Or dispersed to coat whole chamber walls, if that's what you want

03:56 Yes, plasma can 3D print metal, and quite a bit more

If you work in additive manufacturing, powder metallurgy, or aerospace, this is the map.

🔔 Subscribe for more in the "Can Plasma?" series: @gregfridman

#CanPlasma #PlasmaScience #AdditiveManufacturing #3DPrinting #Metal3DPrinting #Powder #Aerospace #Manufacturing #PlasmaSpray

Two viewers (one on YouTube, one on LinkedIn) asked closely related questions about the relationship between ionization degree and plasma temperature. Time for a slightly deeper dive than usual. The answer requires realizing there isn't one temperature. There are four.

⏱️ Chapters

00:00 Two questions, closely related, one YouTube, one LinkedIn

00:03 Both about ionization degree and temperature

00:12 The first question: is ionization related to temperature?

00:38 The second question: why do cold atmospheric plasmas stay cold if electrons are at 1 eV?

01:30 What is temperature? Not what you think.

01:34 A molecule has 4 degrees of freedom

01:47 Rotation

01:51 Vibration

01:56 Translation

02:03 Electronic (my favorite)

02:07 The crowded-room analogy setup

02:29 Rotation: standing still and spinning at near light speed. Nobody notices.

02:52 So rotational energy doesn't heat the gas

02:58 Vibration: jumping in place. You bump your neighbors, but that's it.

03:16 So vibration barely heats the gas either

03:30 Translation: running through the room, bumping into everyone

03:45 THIS is what normal people mean by "temperature"

04:13 Electronic: sit calmly and start screaming. Everyone freaks out.

04:26 This is where non-equilibrium plasmas dump most of their energy

04:37 Highly reactive, highly ionized, and still cold

04:48 How we actually measure: optical emission spectroscopy

04:54 Look at electronic excitation, vibration, and translation temperatures separately

05:04 Translation tells you the gas temperature

05:07 If gas is at 300 K, it's cold plasma. Non-equilibrium. Non-thermal.

05:30 The formal definition: two kinds of plasma

05:34 Equilibrium: ion temp = electron temp

05:38 Both at 1 eV ~ 10,000 K = thermal plasma

05:58 Non-equilibrium plasma = "cold" plasma

06:14 Not that cold. Less than about 10,000 Celsius.

06:21 8,000°C is still "cold" in plasma physics. Doesn't sound cold to me either.

06:40 The theoretical extreme: dump ALL energy into electronic degree

06:59 Result: fully ionized plasma at room temperature

07:24 Theoretically possible. Not very practical.

07:33 In practice: cold plasmas are ionized well below 1%

07:57 As ionization rises, ions start flying, translational energy rises with it

08:11 Three temperatures actually matter: vibrational (chemistry), electronic (ionization), translational (gas temp)

08:38 Purely non-thermal plasmas CAN be fully ionized in principle

08:47 In practice, ionization degree is much lower

08:52 Almost a 10-minute video already. Let me know if you want more depth.

If you're a physics student, plasma engineer, or you got stuck on why plasma physicists talk about multiple temperatures for the same gas, this is the episode.

🔔 Subscribe for more: @gregfridman

#PlasmaScience #PlasmaPhysics #Temperature #Ionization #StatisticalMechanics #Physics #PhysicsEducation

Do you ever wonder exactly how the world is going to go to shit? A short break from plasma to recommend two books that changed how I think about that question. Both by Annie Jacobsen: "Nuclear War: A Scenario" and her follow-up "Biological War: A Scenario."

⏱️ Chapters

00:00 A dark question, and a book recommendation

00:12 Plasma and nuclear engineering are often the same department

00:18 Today: Annie Jacobsen's "Nuclear War: A Scenario"

00:28 Her follow-up: "Biological War: A Scenario"

00:35 Not Hollywood. Not conspiracy theory.

00:43 Deep detail, real interviews, real experts

00:49 Some of the names I recognized from scientific literature

00:56 My father was a Soviet-era nuclear physicist. Extra context for me.

01:04 The structure: one "what if" fiction premise, everything else nonfiction

01:19 The scenario: US gets nuked by an unknown enemy, multiple incoming attacks

01:26 How does the US respond? Russia? Satellites? Real decisions from real history.

01:40 I listened as an audiobook and could barely put it down

01:56 How ridiculously human humans are, once you get the actual details

02:07 The Hollywood image: mushroom cloud, 007 saves the day

02:23 That's not how it actually goes down

02:39 The scary ending: 10,000 years after, 24,000 years after

02:57 How fast it all happens

03:03 The biological war book: I'm halfway through

03:10 How little protection we actually have

03:20 I hate paying taxes. Nobody loves paying taxes.

03:29 But a big chunk of that money goes toward protecting us

03:41 After reading these: take more. 1-2% more, for better protections in place.

04:15 Highly recommend both books

If you liked my previous book recommendation episodes (Ian Banks' Culture series, Asimov's Foundation), this is a continuation of that thread.

🔔 Subscribe: @gregfridman

#Books #BookRecommendation #AnnieJacobsen #NuclearWar #BiologicalWar #Nonfiction #NationalSecurity #Reading

Next in the "Can Plasma?" series: a two-part viewer question about hair. Closer shave? Change your haircut? Change your hair color? And a callback to the ball lightning idea from a previous episode. Short one. Big finding buried in the middle.

⏱️ Chapters

00:00 Today: hair growth and hair color

00:08 The viewer's question in full

00:19 Interesting question, and not a silly one

00:22 What the research actually covers: hair regrowth

00:29 Our work: indirect plasma treating a proprietary liquid, applied to the scalp

00:36 Yes, plasma stimulates hair follicles to regrow hair

00:45 Closer shave? Not really

00:49 The laser hair removal analog

01:04 Plasma as a photocatalyst with additive chemicals for hair reduction

01:18 Or plasma-treated liquid as a surfactant replacement for shaving gel

01:34 Summary: closer shave no, hair removal maybe, hair regrowth yes

01:49 The commercialization problem: this is home use, hard business model for expensive plasma machines

02:06 Can plasma change your hair color?

02:10 Not the way hair dye does

02:13 But gray hair is follicles failing to produce pigment

02:20 Preclinical trials: plasma stimulates pigmentation

02:34 Plasma-treated follicles can regrow darker hair

02:43 Not purple. But silver can go, and darker hair grows back.

03:00 Callback: the viewer asked about ball lightning

03:07 I'm totally into the ball lightning project

03:13 Warning: many have tried to weaponize plasma, it gets tricky

03:32 Starting point: plasma welder, slightly conductive water (a bit of salt)

03:41 You're welding water. Be careful, buddy.

03:44 Wrap-up, questions in the comments

If you're in cosmetics, dermatology, hair loss research, or you saw a "plasma comb" ad and wondered if any of it is real, this is the episode.

🔔 Subscribe for more in the "Can Plasma?" series: @gregfridman

#CanPlasma #PlasmaScience #HairRegrowth #GrayHair #Dermatology #PlasmaMedicine #CosmeticScience #HairLoss

A viewer asked a great foundational question: how ionized does something have to be to be called plasma? Some sources say all electrons stripped from all atoms. Others say one is enough. What's the actual answer? Here's the simple version and the PhD version.

⏱️ Chapters

00:00 The question: how ionized to count as plasma?

00:07 Quick intro, Greg Friedman, plasma bioengineer

00:16 Reading the viewer's question in full

00:24 One source says: strip every electron from every atom

00:31 The viewer's guess: one electron, not all atoms

00:39 Do highly excited atoms count? Is the cutoff 1%? 50%?

00:53 My take on the source: I don't know them, don't need to fight

01:07 The color analogy: what makes something "white"?

01:20 The Pantone system, and why plasma physicists needed the same

01:31 A print shop story from my undergrad

01:52 The formal answer: plasma physicists devised a curve from dark discharges to thermonuclear plasmas

02:17 The simple answer for undergrads and summer high-schoolers

02:23 If you can see it and hear it, it's plasma

02:56 Beyond that, it's books and books of details

03:02 Either adhere to the peer-defined convention or make up your own

03:21 What being a scientist means (adhering to peer conventions)

03:38 So on the "every electron stripped" claim: no thanks

03:49 My working definition: at least one electron stripped from at least one atom

04:17 The density math: 10^19 atoms per cubic centimeter of gas

04:29 1% of that is 10^17 atoms ionized (enormous)

04:38 That degree of organization is huge, and I don't work anywhere near it

04:57 Atmospheric-pressure dielectric barrier discharges: closer to 10^12 ionized/cm³

05:07 7 orders of magnitude lower, so about 0.001% ionization

05:20 That's a strong DBD in air, and that's definitely plasma

05:32 Welding torch: possibly up to ~90% ionization

05:44 Fusion reactors and tokamaks: nearly every molecule stripped

05:55 So 1% is a lot. 50% is huge.

06:04 The correct answer depends on what plasma you're working with

06:08 Fusion: 50% is probably not enough

06:11 Surface treatment: 1% is way too much

06:20 Wrap-up: either the PhD study version, or "see it, hear it"

06:37 Send deeper questions in the comments

If you're a physics student, plasma researcher, or you got into an argument on Reddit about what actually counts as plasma, this is the episode.

🔔 Subscribe for more: @gregfridman

#PlasmaScience #PlasmaPhysics #Ionization #ScienceCommunication #Physics #PhysicsEducation