> What is the technical reason why bitcoin transactions can't scale?

Bitcoin can and has improved its scalability. But there are technical bottlenecks that present tradeoffs between achieving bitcoin's goals and increasing the block size (or shortenting the block time).

I wrote a paper a while back that goes through the exhaustive list of bottlenecks: https://github.com/fresheneesz/bitcoinThroughputAnalysis

> Is it technically infeasible to have a secure network while at the same time having faster transactions, without second layer solutions?

Its absolutely technically feasible to increase the transaction throughput of bitcoin. With technical innovations and hardware improvements, bitcoin's onchain throughput will increase over time. However, its a slow process.

> Can someone help me understand why we are not seeing many changes to bitcoin

Bitcoin's philosophy is slow and steady wins the race basically. The Bitcoin community has been committed to widespread buy in from people around the world before making changes. Proposals are discussed by tons of people, code changes are reviewed by tons of people, and deployments are done in a way that requires both overwhelming support from developers, reviewers, users, and miners. All this is done in as decentralized a way as possible. This process makes it substantially less likely that bugs and mistakes are introduced into the software. It makes the system more solid, secure, and reliable. It builds world trust that the Bitcoin system will continue to be solid and predictable. The trade off is that this process is slow. But its a worthy trade off I think.

> why the 10min transaction times is something which "must" occur in order to maintain security.

The paper I linked above also goes over this. Other people have mentioned it in comments too. Lower block times increase the advantage the last miner to mine a block gets, which increases pressure for miners to centralize. Decentralization of miners is key to bitcoin's blockchain security, and so we want to make sure we're designing the system to be well away from any significant miner centralization pressure.

You can think of this as related to the speed of light. Information can only travel so fast, and this puts a constraint on how quickly distantly separated entities can come to a consensus. In a decentralized network, very few entities are connected directly, and so information musts travel in many hops. This is exascerbated by the fact that your connections aren't chosen at all based on their physically proximity to you. So you have signals that can travel halfway around the world and back a number of times before consensus can be reached. 10 minutes isn't the time it takes to come to a consensus tho. The standard in bitcoin is 6 blocks - so about an hour. Even if block times were 10 seconds, however, the time to come to a consensus in the network (aka the "finalization time" - the time after your transaction has been confirmed after which you can consider it non-reversible) wouldn't be much different.

However, the time relevant to miners that matters is the time it takes for them to get the last mined block so they can start mining on it. If it took 1 minute for blocks to get to miners, that would be 10% of the time available to mine - so miners close to whoever last mined the block would have a 10% advantage. That would be bad centralization pressure. But that would be the same centralization pressure if the blocktime were 1 minute and it took 6 seconds to get to propagate the block to other miners. 6 seconds is not a lot of time for a decentralized network.

> If miners were 100x more powerful (or 100x less powerful) than they are currently, would that change the transaction time?

It would not by the way that the protocol is designed. It targets 10 minutes by adjusting the "difficulty" every two weeks. But if you're asking could we reduce the block time if miners were more powerful? The answer is also no, the block time is related to network propagation, not miner hashpower.