How will the Lunar Gateway go to L2 and L1; how much delta-v is needed?

Question

In a talk to the press which can be heard in the YouTube video NASA Administrator Jim Bridenstine Explains the Lunar Gateway, NASA director Jim Bridenstine says after about 02:40:

It’s also true, because it’s solar-electric propulsion it’s not as big as the international space station. But with solar-electric propulsion, it’s not just going to be in an orbit around the moon, it’s going to actually go to L2 and L1, and give more access to more parts of the moon than ever before.

Question: How will the Lunar Gateway go to L2 and L1, and how much delta-v is required to do so? I have read in other answers here that it will be in a near rectilinear halo orbit (NRHO) around one of those two points, I'm wondering if it will somehow jump between orbits around L1 and L2.

The nature of three body orbits allows for low energy transfers between different (quasi) periodic orbits, so this might certainly be possible with solar-electric propulsion if done very carefully but that might take a long time. (See for example Going from LEO to lunar using only low-thrust ion propulsion - can it be done?) I'm just wondering if this kind of transfer has been worked out in this case, and between L1 and L2 NRHO's in particular.

Background:

• What is a near rectilinear halo orbit?
• Why is a near rectilinear halo orbit proposed for LOP-G (formerly known as Deep Space Gateway?)
• How to find Near Rectilinear Halo Orbits (NHROs)?
• Low Energy Transfer within Earth-Moon system (as-yet unanswered question)
• A longer section of a transcription of his talk can be read here.

Answer 1

Whether or not Jim Bridenstine said it, gateway won't literally go to L1 and L2.

The current plan is to use a Southern L2 NRHO (Near Rectilinear Halo Orbit) which doesn't actually go near the L2 point. If it were a regular Halo Orbit, it would appear to orbit L2, and be closer to L2 than the Moon. A Near (read: Almost) Rectilinear orbit is offset from L2 so much that it's almost a straight up and down line.

I've recently spoken with an orbital dynamics specialist who is working on the Gateway project. They helped me in my research for my own video on this subject How To Land On The Moon (out soon on the smallstars youtube channel.) The person I spoke to made this illustration of where Gateway's orbit will actually be:

Answer 2

There are many ways to get to Gateway's destination NRHO. The transfer that a given spacecraft takes will likely depend on how quickly the vehicle needs to arrive. If it's a cargo vehicle, it can afford to take a long time for less fuel. If it's a crew vehicle, they probably have to go more direct.

I can't really speak to the direct transfers, but I can refer you to some work on low-energy transfers to NRHOs. One study that has been published leverages Ballistic Lunar Transfers (BLTs). In essence, the launch vehicle sends you on a trajectory whose apogee is VERY high. Like, 1.5 million km. At that distance, you can leverage the effects of solar gravity to raise your perigee and/or change your inclination "for free." With a BLT, you use this effect to raise your perigee to the Moon's orbit.

The end result is that, given the launch vehicle has put you on a trajectory with sufficient C3, the spacecraft can complete the transfer and insertion into the NRHO for less than 100 m/s.

Full disclosure: I work for the company I am citing below.

Here's a video that shows this in action (in particular the part between 0:30 and 1:00):

The same company working on the CAPSTONE mission has published conference papers and white papers on BLTs to NRHOs: https://advancedspace.com/blt/. That page gives a quick overview and shows another cool video of a simultaneous rendezvous of three vehicles into the NRHO, each launched at different times. There is a link to the conference paper (no paywall) at the bottom of the page.