This paper has a couple of references to hot-injection synthesis of metal oxide nanoparticles. I highly recommend using hot-injection as it gives you the most control over size, shape, and composition. Once you have the nanoparticles from hot-injection synthesis, you'll need to perform a ligand exchange so the nanoparticles can be dispersed in acetic acid. For my research, I remove oleylamine with diammonium sulfide.
Disperse the particles in hexane (~20 mg/mL) and add an equal volume of NMF with 50 uL aqueous (40-48 wt%) diammonium sulfide / mL NMF. After sonication, the nanoparticles will transfer to the polar phase. Remove the polar phase and wash it with sonication in hexane. Add acetone to the mixture to precipitate the polar phase. Centrifuge and decant the supernatant, disperse the floc with 1-2 mL of NMF and repeat the wash with the antisolvent acetone. Dry the particles with nitrogen and disperse them in the acetic acid. Avoid air exposure of the particles, as they precipitate with air.
If you can't abide using diammonium sulfide which will sulfidize the oxide nanoparticle surfaces, you should be able to make a 0.01-0.1 M solution of acetic acid in NMF for ligand exchange. If you don't have NMF, you can use formamide with capping solution and toluene for the non-polar phase.
Check the procedure in this publication. You can also track back to listed references for detailed experimental procedure.
Dissociative Binding of Carboxylic Acid Ligand on Nanoceria Surface in Aqueous Solution: A Joint In Situ Spectroscopic Characterization and First-Principles Study
Ligands can be omitted during laser ablation sythesis in liquid. Isoelectric point of cerium oxide is in the neutral Ph range, so that acidic condition may contribute to electrostatic stabilisation without using surfactants.
Here you find 2 short, open-access editorials about this method: