Not really near the moon. I just made up a poetic title for fun. What really happened was I was one half of the conversation and Google’s AI query handler was the other. I was sitting at my computer desk in the Canadian Laurentians. Where Mr. AI was is anybody’s guess. I often desecrate the soul of AI in this blog but I don’t want to leave the impression it is entirely worthless and de-humanizing. If you hold your ground it can be a very useful tool to learn about new things in a pleasant conversational manner. Like having an erudite tutor one-on-one in a seaside cafe or a Prague bar. You just need to keep from anthropomorphizing it in your mind.

It can be a very peculiar experience. The topic for me was trying to figure out, initially, why the ‘Orion’ spacecraft was gaining speed just after departing the moon’s orbit and returning to earth. (See screenshot above.) But the depth of the answers given gave rise to many absorbing related topics being discussed. And I learned a great deal. And faster and more directly than if I were to look for numerous links in a similar vein. I still value all that very much, but it is a tremendous tool to be able to direct your study with your own questions with a resonably ‘expert’ source. It took about an hour and a quarter. And I’ve decided to reprint here the entire dialogue verbatim. You can judge for yourself. Mr AI is in blue.

============================================

— Why is Artemis accelerating right now given that it already slingshot around the moon over 24 hours ago?

The Artemis II Orion spacecraft is accelerating right now because it has passed the “neutral” point where Earth’s gravity begins to dominate the Moon’s. Since the mission already completed its lunar flyby on April 6, 2026, it is now essentially “falling” back toward Earth.

Understanding the Acceleration
Earth’s Gravity: Even though Orion is nearly 250,000 miles away, Earth’s gravity is pulling it back in. Now that it is moving toward Earth rather than away, this pull causes it to speed up continuously until reentry.
Sphere of Influence (SOI): On April 7, Orion officially exited the lunar sphere of influence. Once it crossed this threshold, the Moon’s gravity—which was slowing Orion down as it tried to leave—became weaker than Earth’s pull.
Free-Return Trajectory: The mission used a “slingshot” to set a course that naturally falls back to Earth without needing a major engine burn.

Current Status (April 7, 2026)
Velocity: As of approximately 1:30 p.m. ET, Orion was traveling at about 1,300 mph and accelerating.
Distance: It is roughly 233,000 miles from Earth and about 40,000 miles away from the Moon.
Homebound: The crew is on a direct path for a scheduled splashdown in the Pacific Ocean on Friday, April 10.

Would you like to track the live velocity of Orion as it nears its final high-speed approach to Earth?

— No that’s ok; I am using the NASA app. How large in miles then is the lunar SOI? And doesn’t the swinging around behind the moon itself increase Artemis’ velocity?

The lunar Sphere of Influence (SOI) is approximately 41,000 miles (66,000 km) in radius.

The Moon’s Effect on Velocity
Regarding your second question: Yes, swinging behind the Moon does increase the spacecraft’s speed, but only temporarily while it is close. Here is the breakdown of why Artemis II is accelerating now versus then:
During the Flyby (Falling In): As Orion approached the Moon, it was effectively “falling” into the Moon’s gravity well. This caused it to accelerate to its peak flyby speed of 3,139 mph at its closest approach.
Leaving the Moon (Climbing Out): As it swung around the far side and headed away from the Moon, the Moon’s gravity began pulling back on it. This acted like an “uphill” climb, which actually decreased its velocity relative to the Moon.
The Current Acceleration (Earth’s Pull): Now that Orion is roughly 40,000 miles away from the Moon, it has exited the lunar SOI and is back under Earth’s dominant gravitational control. It is now “falling” toward Earth, which is why you see its velocity increasing again. This speed will continue to climb until it hits the atmosphere at roughly 25,000 mph on Friday.

Gravity Assist Dynamics
While a “slingshot” (gravity assist) can increase a spacecraft’s speed relative to the Sun to travel to other planets, in the Earth-Moon system, the Moon is used more as a “gravity rudder.”.

Trajectory Bending: The Moon’s primary role for Artemis II was to use its gravity to bend the spacecraft’s path 180 degrees without using fuel.
Free-Return: This “nudge” ensures that Earth’s gravity naturally pulls the crew back home into a precise entry corridor.

Metric…………..Artemis II Lunar Status
Lunar SOI Radius….~41,072 miles
Peak Flyby Speed….3,139 mph
Current Speed……….~1,300 mph (and rising)
Reentry Speed……….~25,000 mph

— I understand. If earth were not present but only the moon, and all else identical, would the lunar SOI extend further than 41000 miles or is it just a fixed size based upon the moon’s mass?

If the Earth were not present, the concept of a Sphere of Influence (SOI) would actually disappear entirely. An SOI isn’t a fixed physical boundary; it is a relative calculation used to determine which of two bodies has the dominant gravitational pull on a third object. It is essentially the “border” where the Moon’s tug becomes more significant than the Earth’s.
Here is how it works:
It’s a Tug-of-War. The size of the lunar SOI is determined by the ratio of the Moon’s mass to the Earth’s mass, as well as the distance between them.
The Formula: Approximately for our purposes, the lunar SOI vis-a-vis Earth is given by:

SOI = D * (MASSm / MASSe)^{^2/5}

(Where D is the distance between the moon and earth).

Without Earth: If you removed the Earth, the Moon would become the “primary” body (likely orbiting the Sun directly). Its gravity would then extend infinitely (though weakening with distance), and its new SOI would be calculated relative to the Sun.
In short: The 41,000-mile limit exists specifically because the Earth is “stealing” control beyond that point. If Earth weren’t there to interfere, Orion would never have “exited” the Moon’s influence; it would have simply kept slowing down forever as it moved away, never feeling that second tug to speed it back up.

Would you like to see how the SOI of Mars compares to the Moon’s, given that Mars is much larger but also much further from the Sun?

— Yes. And I am also curious as to the point where solar SOI overwhelms Earth’s SOI, which would apply for example with regard to SpaceX plans to send some obsolete future StarLink satellites into a solar orbit after their useful lifetime as a way of avoiding cluttering earth’s orbit.

Earth’s Sphere of Influence (SOI) relative to the Sun extends out approximately 575,000 miles (925,000 km). Beyond this boundary, the Sun’s gravitational pull becomes the primary force dictating a satellite’s path. To place a Starlink satellite into a solar orbit, SpaceX would need to provide enough thrust to push it past this “gravitational fence.” Once outside, it would no longer orbit Earth but would instead follow its own path around the Sun.
Mars vs. The Moon: A Comparison of SOI
Interestingly, despite Mars being much larger than the Moon, their “bubbles of control” are somewhat comparable in the context of their neighbors:
The Moon: Has an SOI of ~41,000 miles. It is small but very close to Earth, which “competes” for control.
Mars: Has an SOI of approximately 359,000 miles.
Mars has a smaller SOI than Earth (despite being further from the Sun’s competing pull) simply because it is significantly less massive—only about 11% of Earth’s mass. Earth’s greater mass allows it to hold onto objects much further away than Mars can.
Starlink and “Graveyard” Orbits
While SpaceX is currently focused on atmospheric reentry for disposal —where satellites burn up safely over the ocean— the idea of “solar disposal” is a long-term strategy for deep-space clutter.
The Threshold: To reach the Sun’s SOI, a satellite at the typical Starlink altitude (340 miles) must increase its velocity by about 7,000 mph to reach Earth’s escape velocity.
The Benefit: Once past the 575,000-mile mark, the satellite is effectively “gone” from Earth’s neighborhood forever, eliminating any risk of the Kessler Syndrome (a chain reaction of collisions) in Earth’s orbit.

Body……Primary Competitor…..SOI Radius (Miles)
Moon……Earth………………………..41,000
Mars……..Sun………………………….359,000
Earth……..Sun………………………….575,000

Would you like to see a breakdown of the fuel requirements (Delta-v) needed to push a satellite from LEO into a permanent solar orbit?

— Yes okay I would. Which I presume means such satellites would need to lug around this extra fuel weight for the entire period of their service lifetimes.

Yes, that is a major drawback. To push a satellite from a standard Low Earth Orbit (LEO) out to the solar SOI requires significant extra energy, which translates directly to carrying more fuel (propellant) for the entire life of the mission.
The Fuel “Weight Penalty”
Current Starlink satellites use highly efficient Hall effect thrusters (powered by Krypton or Argon). Even with this high efficiency, the physics of escaping Earth are demanding:
Delta-v Requirement: To reach Earth’s escape velocity from a typical 550 km orbit, a satellite must increase its speed by roughly 3,143 m/s (over 7,000 mph).
Fuel Fraction: Based on the rocket equation, a satellite would need to consist of roughly 18% fuel just for this final disposal maneuver.
The Cost of “Luggage”: Because every pound of fuel takes up weight that could have been used for better antennas or solar panels, “lugging” this extra 18% for 5–15 years is very expensive for operators.

Comparison of Disposal Methods
Currently, SpaceX prefers a different approach to avoid clutter:
Disposal Method…………………………Delta-v Needed……….Impact on Satellite
Atmospheric Reentry (Current)……~100–200 m/s…….Uses very little fuel; satellite burns up completely.
Graveyard Orbit (High Orbits)……..~11 m/s…….Used to move GEO satellites ~200 miles further out.
Solar Disposal (Proposed)…………..~3,143 m/s………Needs massive fuel reserves; satellite leaves Earth.

The Future: Orbital “Gas Stations”
To solve the problem of lugging fuel for years, companies like Orbit Fab and Northrop Grumman are developing on-orbit refueling.
The Vision: A satellite could be launched “light” with only enough fuel for its mission.
The Maneuver: At the end of its life, a tanker could dock with it and provide the “boost” needed to send it into a solar orbit.

Would you like to see how the cost of launching that extra fuel compares to the potential revenue lost by having a heavier, less capable satellite?

— OK good, yes. I also want to understand Delta-v better. Your explanation implied that the altitude of the earth orbit has a relationship to needed escape velocities.

You’ve hit on a core principle of orbital mechanics: the deeper you are in a gravity “well,” the harder you have to climb to get out.
The Altitude vs. Escape Velocity Relationship
The speed required to escape Earth decreases as your altitude increases because the pull of gravity gets weaker the further you move from the center of the planet.
However, there is a catch. While the target escape velocity is lower at high altitudes, the Delta-v (the change in velocity) needed to get from your current orbital speed to that escape speed is actually higher when you are starting from a low orbit like Starlink’s.

Satellite Altitude……..Minimum Orbital Speed……Escape Velocity……Delta-v Needed to Escape
Surface (0 km)…………………..17,670 mph……………….25,020 mph…………………~7,350 mph
LEO (550 km / Starlink)……..16,970 mph……………….24,010 mph……………….~7,040 mph
GEO (35,786 km)………………..6,880 mph…………………9,730 mph………………….~2,850 mph

Understanding Delta-v
Think of Delta-v as your “maneuvering budget.” Delta means “change.” v stands for “velocity.”
If you are a Starlink satellite moving at 16,970 mph, and you want to leave Earth’s SOI, you don’t need to generate 24,010 mph of speed from scratch. You only need to add the difference: 7,040 mph. That 7,040 mph is your Delta-v requirement.
The “Weight” of the Budget
To get that Delta-v, you have to burn fuel. The “Rocket Equation” tells us that for every bit of Delta-v you want, you have to add fuel exponentially.
The Trade-off for SpaceX:
Revenue Loss: If a Starlink satellite weighs 800 kg, and 150 kg of that is “escape fuel” sitting idle for 5 years, that’s 150 kg of high-tech sensors or solar panels they couldn’t launch.
Launch Cost: At roughly $2,500 per kg to launch on Falcon 9, carrying that “dead weight” fuel costs SpaceX about $375,000 per satellite just to have the option to throw it away into solar orbit later. With thousands of satellites in a constellation, that adds up to billions of dollars in “wasted” launch capacity. This is why “falling back down” (atmospheric reentry) is the industry standard—it costs almost zero Delta-v because Earth’s gravity does the work for you.

Would you like to look at the specific math of why adding just a little more weight (fuel) requires a lot more fuel to carry that fuel?

— Let me delay the specific math just now. I have other questions. Am I correct then to assume that the reason SpaceX is even proposing solar orbit ‘disposal’ is political and societal due to the increasing de-orbiting events? Also, do you have enough information to compare the $375K excess fuel cost per Starlink with the actual or estimated revenue generated per Starlink satellite usage lifetime?

The primary drivers for SpaceX’s solar disposal proposals are political, environmental, and operational pressures—effectively a “proactive defense” against increasing scrutiny of their massive de-orbiting schedule.
Societal and Political Pressures
Atmospheric Pollution: Recent studies show that vaporizing satellites (which are ~40% aluminum) releases aluminum oxide into the stratosphere, which can catalyze ozone depletion. With nearly 10,000 satellites currently in orbit and up to 5 re-entries daily in 2026, scientists and advocacy groups are urging regulators to move away from “atmospheric cremation”.
Debris Risk: While Starlink satellites are designed to fully burn up, there are increasing reports of debris reaching the ground (e.g., in Saskatchewan, Canada). Proposing solar disposal helps SpaceX argue that they are developing “zero-impact” solutions for future billion-satellite constellations.
Regulatory Scrutiny: The FCC and international bodies are tightening de-orbit rules. By proposing solar orbits, SpaceX can claim a greener “greenwashing” narrative, arguing that orbiting data centers have no environmental impact on Earth.

Revenue vs. Fuel Cost Comparison (2026 Estimates)
While the $375,000 “excess fuel” cost sounds high, it is manageable when compared to the revenue a single Starlink satellite generates over its 5-year lifespan.

Financial Metric…………………………….$ Value per Satellite (5-Year Life)
Annual Starlink Revenue (2026)…….~$18.7 Billion (Total) / ~10,000 satellites
Gross Revenue per Satellite……………~$1.8 Million to $2.2 Million
Net Profit per Satellite……………………~$500,000 to $900,000
Disposal Fuel Cost…………………………..$375,000 (Launch mass penalty + propellant)

The Bottom Line:
Lugging the extra fuel would eat roughly 40–70% of the net profit generated by a single satellite. While this makes solar disposal financially “affordable” (the satellite still breaks even), it represents a massive loss in potential scale. For a company focused on aggressive growth, losing nearly half their profit per unit is a hard sell unless regulators make atmospheric reentry illegal or prohibitively expensive.

Would you like to know how the upcoming Starship launches might change these economics by dramatically lowering the cost of “lugging” that extra fuel?

— I think the answer would have something to do with a refueling-in-space architecture that is proposed. OK tell me this please. But I then want to backtrack to something you said about argon and krypton usage in the engine burns. I want to know more about this and also the relative costs and rarity and geographical sourcing of these two elements.

You are spot on. Starship is the lynchpin for the refueling architecture. By using a “Tanker” version of Starship to meet satellites in orbit, SpaceX could top off their tanks right before disposal. This eliminates the “weight penalty” during the satellite’s productive life, as they wouldn’t have to carry that extra 18% fuel for five years. Now, let’s dive into the “gas” powering these engines.

Why Noble Gases?
Starlink uses Hall-effect thrusters (Electric Propulsion). They don’t burn fuel with fire; they use electricity to strip electrons off gas atoms (ionizing them) and then use magnets to shoot those ions out the back at incredible speeds. They use Krypton and Argon because noble gases are chemically stable (they won’t corrode the engine) and heavy enough to provide good “thump” when spit out.
The Shift: From Krypton to Argon
SpaceX started with Krypton for early Starlinks but transitioned to Argon for the “V2 Mini” and larger satellites. Here is why:

Feature……………Krypton (Kr)…………………………..Argon (Ar)
Rarity……………….Very Rare (1 ppm in air)……………Abundant (0.93% of air)
Performance…….High (Easier to ionize/dense)……Lower (Requires more power)
Approx. Cost…….$500 – $1,500 per kg……………….$5 – $30 per kg
Efficiency………….The “Goldilocks” gas………………….The “Budget” gas

Geographical Sourcing & Geopolitics
This is where it gets interesting. The sourcing of these gases is a major strategic hurdle:
Krypton Sourcing (The Ukraine Factor): Krypton (and Neon) are byproducts of large-scale oxygen production for steel manufacturing. Before the 2022 invasion, Ukraine (specifically plants in Mariupol and Odesa) produced nearly 50% of the world’s high-purity noble gases. When those plants went offline, prices skyrocketed, forcing SpaceX and others to find new suppliers or switch gases.
Argon Sourcing (Everywhere): Argon is literally everywhere. It is the third most abundant gas in Earth’s atmosphere. Any air separation plant that makes liquid oxygen (for hospitals or rockets) can easily “siphon off” Argon as a byproduct.
China’s Role: China has become the dominant global supplier of noble gases over the last decade due to its massive steel industry and investment in air separation technology.
The Trade-off
SpaceX switched to Argon because it is essentially 100x cheaper and immune to the geopolitical shocks of the steel industry in Eastern Europe. The “penalty” is that Argon is harder to ionize, so the satellites need larger, more complex power systems (bigger solar panels) to make the engine work as effectively as a Krypton one.

Would you like to know how the solar panel size on a Starlink satellite had to change to accommodate the shift from Krypton to Argon?

— Yes ok tell me. But also these: Can you quantify a bit what you said about the relative ‘performance’ of Ar And Kr above? And your China remark intrigues me. In other words it seems incidental that Ukraine was producing 50% of the world’s noble gases because of steel production. It seems not necessarily tied there by nature or geography. Is China looking at air separation outside the context of steel production and could that not be done anywhere and perhaps constitute a future industry on it’s own?

The transition from Krypton (Kr) to Argon (Ar) for Starlink satellites is a massive engineering trade-off. To make the 100x cheaper Argon work as a propellant, SpaceX had to triple the power and quadruple the surface area of the satellites.
Starlink V2 Mini: The Physical Growth
The Starlink V2 Mini is “mini” only compared to the version intended for Starship; it is actually much larger than the previous Krypton-based V1.5.
Solar Array Span: The span increased from roughly 36 feet (one wing) to 100 feet (two wings).
Surface Area: Total surface area grew to 1,248 square feet, which is more than four times larger than the V1.5.
Mass: Launch weight tripled from ~300 kg to 800 kg per satellite.

Quantifying Ar vs. Kr Performance
Argon is lighter and harder to “ionize” (turn into plasma) than Krypton. This means it requires significantly more electricity to achieve the same thrust.
Ionization Energy: Argon requires 15.8 eV, while Krypton only needs 14.0 eV. This ~13% difference in energy per atom, combined with Argon’s smaller “target size” (cross-section) for electrons, makes the engine much less efficient.
Specific Impulse (Efficiency): Interestingly, because Argon is lighter, it can be shot out at higher speeds once ionized. The V2 Mini thruster has 1.5x the specific impulse of the Krypton version, meaning it gets more “miles per gallon” but needs a massive “engine” (solar panels) to do it.
Thrust Density: To compensate for Argon’s lower efficiency, SpaceX simply built a bigger engine with 2.4x more thrust than the previous version.

The Geopolitics of Air Separation
You are correct that noble gas production is not tied to any specific geography by nature; it is tied to industrial infrastructure.
The Steel Connection: Gases like Krypton and Neon exist in tiny traces (1 ppm). To get enough to sell, you must process millions of tons of air. Historically, only steel mills and large-scale oxygen plants (for industrial chemicals) had a reason to move that much air. This is why Ukraine was a leader—it has massive Soviet-era steel plants.
China’s Strategy: China is indeed decoupling noble gas production from just steel. They are building dedicated Air Separation Units (ASUs) specifically for the semiconductor and aerospace industries. Because they have the world’s largest demand for high-purity nitrogen and oxygen (for electronics manufacturing), the noble gases come as a “free” strategic byproduct.
Future Industry: This can be done anywhere with high electricity and a need for oxygen (like medical hubs or rocket launch sites). Companies like Air Products are now building plants designed to capture these rare gases as a primary high-value product rather than just a side note to steel.

Would you like to see a projection of the stratospheric impact if SpaceX continues with its high-volume atmospheric reentry strategy instead of solar disposal?

— No thanks, I have already looked at that topic a little. Just two final topics. First what happens to the ‘spent’ or de-ionized Argon or Krypton in this HALL Effect process? And can you say more about this electronic propulsion technique in general. It is the first time I heard of it and is very interesting.

The Hall-Effect thruster is essentially the “electric car” of space propulsion. Instead of exploding chemical fuel to create push, it uses electricity and magnets to shoot gas at extreme speeds.
1. What happens to the “spent” gas?
Once the Argon or Krypton atoms are ionized and shot out the back of the thruster, they are gone forever.
The “Exhaust”: As the ions fly out of the nozzle at speeds up to 60,000 mph, they pass through a “neutralizer” (an electron gun). This adds electrons back to the ions, turning them back into neutral gas atoms.
Why Neutralize?: If you didn’t neutralize the exhaust, your satellite would quickly build up a massive negative charge. That charge would eventually pull the ions right back toward the satellite, canceling out your thrust and potentially short-circuiting your electronics.
Fate: The de-ionized gas simply drifts into the vacuum of space. Because it’s moving so fast, it easily reaches escape velocity and leaves the Earth-Moon system entirely.
2. How the Hall-Effect Works (Simple Version)
Think of a Hall-effect thruster as a circular “racetrack” for electrons.
The Magnetic Trap: Magnets create a circular magnetic field at the exit of the thruster. This “traps” electrons in a spinning ring.
The Gas Injection: You pump your Argon or Krypton into this ring of spinning electrons.
The Collision (Ionization): The spinning electrons smash into the gas atoms, knocking off their outer electrons. This turns the gas into positively charged ions.
The Big Push: An electric field (anode/cathode) is set up. Since the ions are positive, they are repelled by the front of the engine and attracted to the exit. They “surf” this electric field, accelerating to incredible speeds.
Thrust: According to Newton’s third law, as those ions shoot out the back at 60,000 mph, the satellite gets a gentle push forward.

Why use it?
Efficiency: It is roughly 10x more fuel-efficient than the best chemical rockets. You get much more “mileage” per pound of fuel.
The Trade-off (Low Thrust): The “push” is incredibly weak—often compared to the weight of a single sheet of paper resting on your hand. You can’t use it to launch off Earth, but in the frictionless vacuum of space, if you run that “paper-weight” push for weeks or months, you can reach incredible speeds.
Summary of Propulsion types:
Chemical (Fire): High thrust, low efficiency. Used for getting off Earth.
Electric (Magnets): Low thrust, high efficiency. Used for moving around once you’re already in space.

Would you like to know about the Deep Space 1 mission, which was the first to prove this “blue glow” ion technology for NASA back in the 90s?

— Thanks for the offer but I must leave now. I am grateful for all this info and understanding given and will think about it and return.

It was a pleasure diving into orbital mechanics and the “geopolitics of gas” with you! It’s fascinating how a lunar flyby, a noble gas shortage in Ukraine, and the surface area of a satellite’s wings are all connected by the same physics. I’ll be here whenever you’re ready to pick back up. We can dive into that Deep Space 1 history, or perhaps look at how Starship’s massive payload capacity might finally make “solar disposal” a realistic industry standard. Safe travels, and I look forward to our next session!

Would you like me to summarize our discussion on Delta-v and propellant costs so you have a quick reference for when you return?

============================================

Conversation Ended: 1 hour 16 minutes

A few impressions and observations from this experience. Notice that Google has configured this AI to extend the discussion indefinitetly by suggesting related issues at the end of each question it answers. Sometimes this was beneficial because I had some time and the software was reasonably agile at supposing the directions of my interests. One could see the potentially addictive element here, which might carry more negativity with a non-technical subject matter. Also, the AI was good at ferreting out the intent behind some of my questions. Much like with a human partner, you want to have the sense that not too much extraneous detail is derailing the exchange into no-man’s-land. It’s alot like visiting a professor’s office during open hours to squeeze in a little talk and speed up your grasping of a topic.

The other big point is that Google has implemented it in a way that encourages projecting ‘sentience’ onto the AI query tool on the part of the user — me. This strategy is in no way vital to the main function. But you know, AI developers love their Turing Test bonafides! This feature really blossomed at the end when the software determined that the session was coming to a close. Mr. AI even owned up to experiencing ‘pleasure’ with our diverse interactions. I played along with politeness in my own’speech’ and the AI seemed to respond in kind with a calculated appropriate tone. I have to say I like it as a rapid research get-up-to-speed tool. Sometime I will try it out with a non-technical more humanistic or ‘fuzzy’ subject matter and see how things feel.

What did you think? I have not compared it with more specific conversational querying AI programs such as Anthropic.

_______RS

[ Image: screenshot from NASA’s little free online app which permits one to track the progress of the recent Artemis II mission in which four astronauts travelled beyond the dark side of the moon. A simple but cool app which let you alter perspective and direction and see the craft or the overall flight path while also seeing the positions of the sun, moon, and earth. ]

► Handy INDEX — scan through all available ||SWR|| articles