Thank you for sharing the command and output. I think I see why this issue is occurring. As noted in the article above for a single-node cluster, S2D will automatically use 2-way mirroring when creating a volume. As their will be two copies of the same data, the footprint of the volume will be two times the size of the useable space. In order to create a 2TB volume, you would need 4TB of free space. Unfortunately, your storage pool only has 3.49TB of space. This is why the volume creation appears to be failing.
Typically, we recommend always reserving some free space in the pool in case a drive was to fail so that S2D could automatically repair the volume. The amount of space to reserve would be based on the size of the drives that were being used in the pool itself. Using your drives from above, we would typically recommend reserving 894.25GB, the size of one of the drives. If you were to do that in this cluster, that would only leave you with 2,682.75GB of free space to create your volume. And since S2D uses 2-way mirroring for volumes on a single-node cluster, your maximum volume size would be half of that, or 1,341.375 GB.
Reserving free space is optional. If you don't reserve the free space, it just means that S2D won't be able to automatically repair the volume in case of a drive failure until that drive has been replaced. If you decide not to reserve any free space, then the maximum volume size you could create would be approximately 1.745TB (1,788.5GB). Keep in mind that due to the way the calculations are made at the byte level, you may still have to adjust that number down slightly.
Hope this helps!