Why is TE mode preferred to TM mode in the optical waveguide in silicon photonics? I recently started learning about silicon photonics, focusing on its usage in the high-speed interconnect.
Most papers I've been reading so far use optical waveguides choosing TE00 as the dominant mode, and I have not seen any papers or references choosing TM00 as its dominant mode.
For instance, attached is the snapshot from "Silicon Photonics Design from Devices to Systems" by Lukas Chrostowski.

In this figure, if we choose slab thickness of ~240 nm, we can see that two main modes exist, TE00 and TM00, and TE00 has a higher effective refractive index (ERI) than the TM00 mode.
I understand that the ERI can vary depending on the geometry of the waveguide. Still, even so, the author chooses TE00 as the dominant mode for every single example throughout the book. This is just a tip of the iceberg. As mentioned above, every reference I've been reading uses TE00 as the dominant mode without detailed explanations.
In summary, I have the following questions:
Q1: Researchers often choose TE00 as the dominant propagation mode due to its high ERI. My first question is, what is the advantage of having an ERI? Is it related to slowing down the propagation velocity? If so, what is the benefit of slowing down the propagation velocity?
Q2: If it is not due to the ERI, what is the advantage of choosing TE00 over TM00?
I would appreciate any reasonable explanations for my question.
 A: Great Question! I also had this question in mind when i started modelling the devices/waveguides/couplers in Lumerical and studying optical fibers. I've related the emphasis on TE mode with some goals for most applications of the desired silicon devices.
Dispersion & polarization
Propogating lowest TE mode largely maintains the polarization. Modes travel at different speeds, alowing more than one will result in dispersion and distortion in pulse shape.
Higher Bandwidth
The higher ERI of TE00 mode results in its group velocity lower than the TM00 mode. The lower group velocity means that it is less affected by material dispersion, making it more suitable for high-speed data transmission.
Lower propogation loss & smaller waveguide size
Another aspect of having a high ERI is that it enables a stronger confinement of the electric field in the core of the waveguide, which results in a smaller mode size and lower propagation loss. This means that more of the optical power is transmitted through the waveguide rather than being absorbed or scattered. Look at the graph you posted, The lower bound of effective index is cladding's refractive index and upper bound is refractive index of core. You can see that More of the optical power of TM00 is in cladding that we don't want.
Fabrication Complexity
Now look at graph that you've posted, if you change the waveguide size, TM and higher order modes are more effected than TE00. Hence you can say that TM mode has a higher sensitivity to waveguide geometry variations, which can cause a larger propagation loss and a lower tolerance to fabrication errors. The TM mode also has a more complex field distribution that can make it more difficult to efficiently couple to other optical devices.
Easier coupling to other optical devices
Because of its simpler field distribution, the TE00 mode is generally easier to couple to other optical devices, such as single-mode fibers, which is important in practical applications.
Situations where TM mode may be preferred
When working with on-chip devices that already generate a magnetic field, or when working with magnetic materials. Also when coupling a light from free-space, TM mode is generally easier because it's a stronger dipole mode.
