the purpose of this study is the realization of p-type silicon detectors of very high resistivity. the well-known process of li ion drifting compensates for the acceptor impurities in the p-type si of low resistivity used for the manufacturing of si(li) detector becomes useless. previous si(li) detectors present a problem of utilization and storage at low temperature. the obtainment of silicon monocrystals having a resistivity reaching 130.000 omega cm allows to use them directly in detection of nuclear radiations without compensating them with lithium. in order to familiarize with the techniques of detector manufacture in general, we have realized si(li) detectors. afterwards we began the research itself by a study and realization of a great number and different types of detectors from p-type silicon having a resistivity varying from 40.000 to 100.000 omega cm. the surface barrier detectors (sbd) elaborated after an investigation of different chemical treatments present electrical characteristics and dead layer better than si(li) detectors at time t=0. however they present a stability problem. a more profound investigation has to be undertaken in the future to improve the contacts and the surface treatment which are the principal causes of this instability. pin type detectors made by implantation technique gave way to very promising results in nuclear detection. the resolutions which are obtained on the peaks of energy above 20 kev, are much comparable to those obtained with the si(li) detectors. (for example, we have obtained a resolution of 540 ev on the peak of 59,6 kev