I will preface my comments with the following.
In 1999, as a project course in University, I built a alkaline fuel cell from scratch. It worked but had a very low power density. The electrolyte was novel but the electrodes being fluidized beds couldn’t handle high gas flow rates needed for high power densities despite the massive active areas of the cathode and anode electrodes.
This project resulted in me getting my first job as a newly graduated mechanical engineer at Global Thermoelectric, the world leaders in Solid Oxide Fuel Cell (SOFC) technology. I was hired 3 months before I finished school so had no stress about finding a job. I was the 16th person in fuel cells and watched the company grow to about 200 with the tech bubble. I was a fuel cell stack design engineer right from the start, I was working on the cutting edge of the cutting edge.
My first patent was the fuel cell stack compression system for the fuel cells in the trunk on that BMW shown in this photo. I probably was part of hand assembling the very fuel cell stacks in this photo.
https://www.autointell.com/News-2001/February-2001/February-21-02-p3.htm
I learned a lot, built some cool stuff, got several patents (I currently have 33 issued patents), and made some amazing friends with some very talented people.
The company had a board that didn’t understand the technology and sold the company to FuelCell Energy (FCE) for less money than we had in cash. FCE spun us out into a consortium with The Gas Research Institute, Colorado School of Mines, and some group in Utah. The new co was called Versa Power Systems, I worked for 3 companies in a year and sat at the same desk. I got head hunted regularly and eventually after 5 years left for a very lucrative job in the oil patch. Financially I made the right choice.
Versa is now back under full control of FCE. They have done some neat projects for DARPA, Cummins, and others over the years.
So here is my take on the OPs questions.
Hydrogen has a really poor energy density, even compressed. If you liquify it, you consume 1/2 - 2/3 of the usable energy in the hydrogen getting it there so this is not an efficient option. Anything hydrogen powered will have a terrible range.
Hydrocarbons such as natural gas (CH4), propane (C3H8), etc have far better energy densities and can be compressed efficiently. This is the right type of fuel to use for transportation.
PEM fuel cells (Ballard) can only run on hydrogen unless you reform the hydrocarbons in the balance of plant which can be done. These fuel cells run at about 80°C so they have an issue with dealing with the water in the exhaust stream on the cells and this kills power density (basically it makes part of the active area of the electrolyte ineffective).
SOFC fuel cells run at 600°C - 800°C (there is a temperature gradient across the cell, especially if you are doing on cell reforming). The higher temps lead to accelerated corrosion and degradation in performance. These can run on any hydrocarbon fuel, CO is a fuel, and of course you can run them on hydrogen but thermal management can be an issue that needs to be mitigated. SOFCs can also be run backwards to electrolyze water back into hydrogen and oxygen. This has an interesting application for spacecraft and other applications. During the night you can run a drone or whatever off of the fuel cell, during the day you can run your drone or whatever off of a solar panel that is large enough to run the fuel cell backwards to replenish your fuel for the night.
The way I see things is this...
Once the hoopla and BS is settled, the right technology will be used in the right applications. For example, SOFCs will be used in auxiliary power units in big trucks because they can run on the diesel tanks already on the trucks.
PEM cells have applications to replace batteries in many applications (I had an opportunity to join Angstrom Power when it kicked off). These guys made fuel cell powered cell phones and power tool batteries. Camping generators are another good application for PEM cells and there was even a Coleman product at one point.
FCE has built large backup generation stations for hospitals using their molten carbonate fuel cell technology. I think SOFCs would have been better.
Internal combustion engines are NOT going away. They are the right tool for the job in many applications and a fuel cell can’t replace them.
There are lots of ways to do stuff and there is not one silver bullet technology that does it all. As an example, Diesel engines are spectacular for over the road transport and heavy equipment. You won’t find a diesel powered lawn mower or weed trimmer. You can get electric versions of both, even cordless. These are examples of where a PEM fuel cell could replace the batteries but then you would need a fuel source, and a hydrogen infrastructure to support it.
I will get off my soap box now. Sorry to bore you all.
Here is a picture of a 25 year old me testing one of my new fuel cell stack designs (it had the highest power density of any stack we had built up until that time)...
A bunch of my friends left to go to Bloom and other places. My old boss is the VP of Engineering at Bloom.