A Search for ``Failed Clusters'' of Galaxies

We describe a search for a new type of object-large clouds of hot gas with no visible galaxies-which we call failed clusters of galaxies. We calculate the expected X-ray luminosity, temperature, and angular diameter of such objects as a function of total cloud mass and convert the results to expected X-ray fluxes from failed clusters at different redshifts. Using the Einstein Imaging Proportional Counter (IPC) database, we establish a strategy to search for candidate failed clusters. From this initial screening of 1435 IPC fields, 17 candidates are selected for more detailed analysis, which indicates that 10 of these are very probably extended X-ray sources. Optical follow-up on the 10 prime candidates finds eight clusters of galaxies (including six reported for the first time in this paper), one stellar identification, and one without an obvious optical counterpart (the candidate with the weakest evidence for X-ray extent). Investigation of several candidates with less evidence for X-ray extent yields two additional new clusters of galaxies. A conservative comparison of our results with the Einstein Extended Medium Sensitivity Survey demonstrates that failed clusters are at most one-sixth as common as clusters of galaxies (98.3% confidence level). Therefore we conclude that failed clusters are a relatively unimportant contributor to the mass density of the universe. Our inability to find failed clusters is consistent with the hierarchical clustering scenario for the formation of galaxies and clusters.