Minimum Quantity Lubrication (MQL)-based machining process has many merits over not only conventional flood cooling machining but also dry machining. These merits include cost reduction, industrial hygiene, reduction of air-bourn particles etc.. However, few disadvantages make the MQL-based machining process impractical to be adopted in many industrial production settings. More specifically, the cutting speed in a typical MQL process is restricted because the MQL process does not allow efficient cooling of cutting tools. The amount of oil used in MQL processes is barely enough for less aggressive machining conditions and does not effectively cool the cutting tool essential in more aggressive cutting conditions. At high cutting speeds, the oil will simply evaporate or disintegrate as soon as the oil droplets strike the tools already heated to high cutting temperatures. In this paper, new nano-platelet-enhanced MQL oils are developed to mitigate this major deficiency of MQL process. In particular, exfoliated Graphite nano-platelet (xGnP) and XGS Hexagonal Boron Nitride (hBN) produced by XG Science, Inc. (Lansing, Michigan) were mixed into a typical vegetable oil used in MQL process. The process enhancement by these nano-platelets is possible because these nano-platelets in a multiple-layer scheme can readily slide to provide additional lubricity. When the mixture of oil and nano-platelets are applied, the nano-platelets provide additional lubricity even after the oil droplets have been disintegrated during high speed machining. Thus, the enhancement achieved by adding nano-platelets allows us to expand the processing envelope of MQL. In this paper, not only the comparison between xGnP and XGS hBN but also multiple combinations of nano-platelet-enhanced MQL oils with varying contents have been studied with a ball-on-disc set-up in a reciprocating motion in order to study the friction and wear characteristics of these lubricants. More importantly, MQL-based ball-milling experiment has been conducted with these mixtures, which show very promising results for future MQL research.