Has anyone ever seen buried expansion loops used?

A client is keen to use expansion loops that are buried in unconsolidated soft soil such as expanded clay in combination with foam, and then a protective soil layer on top. They want to minimize the axial compressive forces in the pipeline to prevent upheaval buckling by allowing it to displace at expansion loops.

I have never heard about buried expansion before.
Because there're no thermal effect for buried pipeline except fluid temperature. It useually close to ambient temperature.

If you really need expansion loop because of thermal expansion of piping you should provide expansion loop above ground by route pipe up to air for that location only.

Or you can use result of CAESAR II to check and confirm effect of thermal expansion.

Note: My experience is less than 7 years.

But one of my ex-colleague had used buried expansion loops in GASCO, Abu Dhabi project. But I am also not sure about this.

If tendency of upheval buckling has been confirmed, reputed oil companies in their design practices accept approcahes like relooking at operating parameters influencing the axial comp force( thickness, temp,), INCREASED FLEXIBILITY( loops, offsets, zig zag routing etc)and Addition of weights in the form of rock.

Yes, buried expansion loops have been used for hot buried pipelines. Similarly, for hot onshore (gas) lines, long radius buried bends have been installed prior to the above ground transition to scraper traps on either end of the pipelines to reduce the trust load on the scraper traps (i.e. the buried horizontal bend takes allot of the thermal expansion stress/pipe displacement. Similarly, contractors have "snaked" pipelines in the ditch to provide additional mechanism to relieve thermal expansion and contraction stresses/displacements.

In addition, you may want to look into strain based design approaches that recognize the nonlinearity of the pipeline steel stress-strain curve to absorb thermally induced stresses for high-pressure-high-temperature (HPHT) buried pipelines. However, for this design approach, CAESAR II cannot be used since CEASAR II is based on linear elastic design principles and small strain theory.

If a horizontal expansion loop were buried, (rather than poking back up above ground, then diving back below), wouldn't it be better supported by a gravel bed in the ditch than foam and clay?

Seems the gravel would allow movement better over the long run than the (cracking) foam and (settling, semi-rigid) clay.