Battery and charging technology. It sucks. On many fronts.
Let's me take a step back in time and talk about a different domain.
I've been flying electric powered RC planes and helicopters since the early 80's. Back then it was very rare to see people at flying fields with electrics. I was usually the only one at a field with 20 to 30 flyers. In fact, I used to design and manufacture motor controllers because the available ESC's were shit. The point is, it was "the early years".
All we could use back then was Nickel Cadmium (NiCd) cells. They could deliver the necessary current, yet imposed a size and weight penalty. My most powerful plane used a pack consisting of 27 NiCd cells. It was a rocketship and could go straight up at an impressive rate, but the thing was heavy (don't remember, I think it was about 8 lbs).
Nickel-Metal Hydride (NiMh) came along and looked promising but didn't really make a huge difference, certainly not for high current, high performance applications --which is most of what I was interested in.
And then we got Lithium-Polymer (LiPo). This was a technology that delivered (just guessing) twice the volumetric energy density, which meant a battery half the volume and half the weight of the old NiCd pack delivered substantially more energy while being able to supply high currents. This is when my planes went from 27 NiCd cells to a LiPo pack with just six cells. My helicopters use one or two of the same packs. And, of course, LiPo's made drones possible due to the same energy/weight ratio.
Back to cars.
This is it. Energy storage technology with twice the volumetric energy density is where the inflection point will be. Same metrics, half the battery pack volume and mass, two to four times the energy storage capacity.
And yet it doesn't end there. There are two more factors. Cost and charging.
Cost should be self-evident. The battery pack represents a massive portion of the COGS of an electric vehicle. This needs to be cut in half and, eventually, half again.
Charging is the elephant in the room. This is particularly evident during emergencies. I still remember when fires here in CA caused massive problems for electric vehicle owners. One of the worse things one could face as a parent is having an emergency and realizing that your vehicles are range limited and impossible to charge.
And so this key differentiator is both an internal and external factor.
I think the external is portion is easy to understand: The installed based of gas stations dwarfs the installed base of electric charging stations. Not only that, even if we had the same number of available electric stations, the realities of charging are not in favor of electric vehicles. This, once again, has two elements to it.
First, batteries take a long time to charge. If we establish this at 10x fuel filling time (3 minutes for gasoline, 30 minutes for charging), this means you need 10x the charging systems per station in order to service the same number of vehicles per unit time when compared to a typical gas station. In most places this is impossible.
Even if you could install 10x the charging systems in order to be able to service the same number of vehicles, you now face the next --and very serious-- problem: Energy demand.
Stated in the simplest possible terms: A rapid charging network with enough capacity to service a non-trivial number of electric vehicles would require an equally non-trivial amount of energy production capacity increase.
The only way I see a path to deliver this would be nuclear energy. In a place like the US you would probably have to build somewhere in the order of twenty new nuclear power plants distributed across the nation. Each country in Europe would likely need a few. Asia, a bunch of them.
Here's where the electric equation collides with reality: Transition to electric vehicles today, with current technology, and face the reality that we might actually produce far more pollution due to the massive step change in energy requirement.
Distributed solar network? Beyond massive, whatever that means. What's the energy, pollution and CO2 footprint of what it would take to manufacture, I don't know, 10x the solar panels we make today? Not to mention the resource utilization, mining and environmental damage this might cause. Likely not a solution. I think nuclear is the only solution. And, at least in the US, it could take thirty years to build just one nuclear plant; with twenty being almost unimaginable. If we started today the energy infrastructure would not be there until 2050. We should have gotten serious about nuclear energy a few decades ago.
This simple analysis tells me that the key differentiator (and the missing link to achieve mass transition to electric vehicles) has got to be something fundamentally different from the battery technology we use today. I think this means some kind of a fuel cell-type technology where recharging can be delivered in a few minutes through the exchange or replacement of a consumable/recyclable/rechargeable liquid.
I am not sure what else would make sense.
And, of course, all of this has to be evaluated on the basis of whether or not we are actually making the world a better place. It is easy to think that electrics make things better while not realizing that electrics at scale --done wrong-- could actually bring forth an ecological disaster the likes of which we have not seen yet.
Again, think 1.4 billion electric vehicles, don't think about your Tesla on your driveway. You have the ability to have that vehicle on your driveway because the source and nature of the energy it requires to operate doesn't quite move the needle, in terms of local or global scale. Also, the origins and source of that energy can remain, shall we say, conveniently ignored.
I am NOT down on electrics. We were ready to buy a couple of Teslas a few years back. When the California fires happened and we saw what was going on with Teslas we had to think things through. Living in fire and earthquake country you have to be aware of these things. We ultimately decided to wait until the infrastructure matured. Not the vehicles, the infrastructure. We want electric vehicles, but they cannot impose restrictions we don't currently have on their usage. That's our metric. Others are free to develop their own.
Which brings me to what I do not think is a key differentiator/technology for electrics to be successful: Self driving.
In my opinion this is a solution looking for a problem.
The evidence is simple: There are somewhere in the order of 1.4 billion cars on the road. People are driving them around every day. No problem to fix. For us this represents exactly 0% of the many variables involved in making a purchasing decision. It isn't important to most people (1.4 billion vehicles without it) and I don't think this is what will compel mass transition to a technology that currently has serious infrastructure issues.