We are looking at a 40-year-old iffy, obscure paper here. Since when POE causes engine deposits? Who knows what went wrong in that paper. They didn't use the proper antioxidant? They used a very unstable POE? More importantly who cares?
Tom NJ is right on that there are many oil-deposit tests and they may or may not correlate with actual engine deposits. TEOST 33C was recently shown not to correlate with turbocharger deposits at all.
It turns out that TEOST 33C test is useless for judging turbocharger protection. Antioxidant is the key, along with the VII (or the lack of it) and dispersant. Moly has no effect. However, I think TEOST 33C may be better correlated with intake-valve deposits (IVD) in turbo gasoline...
www.bobistheoilguy.com
Regarding moly it's interesting that moly is not what causes the deposits in TEOST 33C but it serves as a catalyst to form deposits. This is why SAE 0W-16 and SAE 0W-20 are exempt from TEOST 33C in ILSAC, as Japanese OEM's like to load them with high moly content, such as 600 ppm Mo or above, and such high-moly oils would fail TEOST 33C, despite performing very cleanly in an actual engine.
TEOST MHT is no longer used as far as I know.
Both TEOST 33C and TEOST MHT use a catalyst (a different type in each) to induce deposit formation. This is probably why it is so hard to correlate them with actual engine deposits. They may be more useful for screening purposes when designing oil blends.
buster is right that the additives, detergents, dispersants, and VII also play a major role.
There is some truth to some extent that higher Noack reduces deposits as the oil evaporates faster before it can deposit. Thinner oils also have lower aniline points and better solvency. However, if higher Noack also means inferior base-oil quality, this could increase the deposits.
As for the polybutylene, it does evaporate extremely cleanly, without leaving deposits. However, it looks like it's only available in very high viscosities, such as KV100 ~ 20 cSt; therefore, you would only see it in some monograde etc. applications. It is used as a replacement for bright stocks—high-viscosity Group I base stocks, as these are becoming less and less available as solvent-refining to make Group I base oil is being abandoned.
Coming back to POE, Valvoline Premium Blue Restore is specifically designed as a maintenance oil to clean the carbon deposits around the piston rings in one oil-change interval. Its base oil has 62.5% POE.
Here is the Valvoline Premium Blue Restore formulation.
Formula #4 is what is sold commercially.
Priolube™ 1973: an 8.00 cSt POE by Croda
Synesstic™ 12: a 12.4 cSt AN by ExxonMobil
D3495L: a detergent–dispersant–inhibitor (DDI) package by Infineum
PX-3871: a mixed alkyl borate ester additive for dispersancy, antiwear, and friction modification (antifriction) by Dorf Ketal
It's unlike any other oil. Valvoline Premium Blue Restore has:
Base-oil composition: 62.5% (50/(50+10+15+5)) POE (ester), 25.0% ((15+5)/(50+10+15+5)) PAO, and 12.5% (10/(50+10+15+5)) AN (alkylated naphthalene)
No viscosity-index improver (VII) at all—a monograde oil
Standard Valvoline Premium Blue Synthetic HDEO additive package (20% of the finished oil)
Therefore, unlike what that old paper reported, POE, with its extremely high solvency and extremely low aniline point, does a better deposit-cleaning job than any other base oil.
Euro Mobil 1 oils (FS and ESP varieties) also use POE for better cleaning of diesel engines as well as meeting the severely extended Euro oil-drain intervals.