The use of surfactants for control of specific aspects of the VPE growth process is beginning to be studied for both the elemental and III/V semiconductors. The objective is to change the characteristics of the material grown epitaxially by the addition of a surfactant during growth. Most reported surfactant effects for semiconductors relate to some detail of the morphology of the growing films. For ordered semiconductor alloys the effects can be much more dramatic, including major changes in the electrical and optical properties. Since the bandgap energy is dependent on the microscopic arrangement of the atoms in an alloy with a fixed composition, the change in order parameter induced by the surfactant translates into a marked change in the bandgap energy. This paper presents the results of a study of the effects of n-type (Te and Si), p-type (Zn), and isoelectronic (Sb) dopants on the ordering process in GaInP grown by OMVPE. All of the dopants studied were found to decrease or eliminate ordering; however, the mechanisms are quite different. The donor Te apparently affects the adatom attachment kinetics at steps on the (001) surface, a surfactant effect. On the other hand the donor Si was found to decrease the degree of order by an entirely different mechanism, attributed to an increase in the Ga and In diffusion coefficients in the bulk. It apparently does not involve the surface. Disordering due to the acceptor Zn was found to occur by the same mechanism. The isoelectronic impurity Sb is found to act as a surfactant and to decrease the order parameter by changing the surface reconstruction, eliminating the [ -110]-P dimers that provide the thermodynamic driving force for formation of the CuPt structure during growth.surface by increasing the temperature or decreasing the partial pressure of the group V precursor during OMVPE growth. 3,4 Although the driving force for ordering is understood, the mechanism remains unknown; however, several speculative models have been proposed. 5 Besides the known role of surface reconstruction, surface steps may also be important factors in the ordering process. For example, [110] steps are observed to assist the ordering process but [ -110]steps retard ordering. 6 Kinetic effects at step edges apparently affect the ordering process under certain growth conditions. One of the factors having a strong effect on ordering is doping. Several studies in GaInP have demonstrated a connection between ordering and n-type 7-11 or p-type 12-18 dopant concentration. The results show that a drastic decrease in ordering is caused by intro-