This thesis is the summary of my work on the µ-reactor platform. The concept of µ-reactors is presented and some of the experimental challenges are outlined. The various experimental issues regarding the platform are discussed and the actual implementation of three generations of the setup is described in detail. Since heating and temperature measurement is an extremely important point in heterogeneous catalysis an entire chapter is dedicated to this subject. Three different types of heaters have been implemented and tested both for repeatability and homogeneity of the heating as well as the maximal achievable temperature. The currently best solution for heating and measuring temperature is to use several integrated Pt-strips as ohmic heaters and a separate 4-point Pt meander structure as a resistance thermometer. It is shown how it is possible to quantify the available amount of active catalyst surface area by reacting off an adsorbed layer of oxygen with CO. This procedure can be performed at temperatures low enough that sintering of Pt nanoparticles is not an issue. Some results from the reactors are presented. In particular an unexpected oscillation phenomenon of CO-oxidation on Pt nanoparticles are presented in detail. The sensitivity of the reactors are currently being investigated with CO oxidation on Pt thin films as a test reaction, and the results so far are presented. We have at this point shown that we are able to reach full conversion with a catalyst area of 38 µm2 with a turn over frequency of approximately 105 site-1 s-1. The apparent activation energy for all sizes of thin films is found to be ∼ 1.2eV. We expect to be able to go to even lower amounts of Pt and corresponding higher turn over frequencies by increasing the temperature further. Various methods of gaining insight into the reactor besides the normal quadropole mass spectrometer are discussed. In particular a chapter is devoted to the application of a Time-of-flight mass spectrometer on the setup. Finally the thesis contains an outlook of the possible future of the setup, and several suggestions for improvements are discussed.