Dust particulates immersed in a quasineutral plasma can emit electrons in several important applications. Once electron emission becomes strong enough, the dust enters the positively charged regime where the conventional Orbital-Motion-Limited (OML) theory can break down due to potential well effects on trapped electrons. A minimal modification of the trapped-passing boundary approximation in the so-called OML + approach is shown to accurately predict the dust charge and heat collection flux for a wide range of dust size and temperature.PACS numbers: 52.25. Dg, 52.27.Lw, The problem of the charging of a solid body in a plasma has a long history and various applications ranging from probes and spacecraft to planet formation to dusty plasmas in laboratory and space [1]. The body collects plasma particles and is often negatively charged owing to the higher electron mobility. In many instances, however, the body can emit electrons (via thermionic emission, photoemission and secondary electron emission) and become positively charged. [9,10]. Experiments showing the formation of ordered structures with positively charged dust are reported in laboratory [11,12] and in microgravity [13]. Since charging is governed by the characteristic length of the body relative to the plasma Debye length or the electron gyroradius, there is no conceptual difference between the examples above and in what follows we will use the term dust broadly.A charging theory is a necessary ingredient of any model of dust transport and destruction/survival in a plasma [4,6,7,9,[14][15][16][17][18][19]. It calculates the dust charge/potential, momentum and heat collection due to the dust-plasma interaction. The most widely used charging theory is the Orbital-Motion-Limited (OML) theory [20], which leads to a simple nonlinear equation for the dust potential.In this Letter, we use PIC simulations and theoretical analysis to show that OML can become inapplicable in the positively charged regime. It can completely miss the transition between negatively and positively charged dust (thus predicting a positive dust potential when simulations show a negative dust potential) and overestimates the power collected by the grain (up to a factor of 2 for the cases considered). This is due to the development of a non-monotonic potential (a potential well) near the grain. The fact that a potential well can exist near an electron emitting body is known [3,21,22]. However, this is the first study that illustrates the breakdown of OML in this regime. Moreover, this Letter presents a revised charging theory which is as simple as OML, recovers OML in the appropriate limits, but remains accurate when potential well effects are important.We study the charging of a spherical dust grain of radius r d at rest in a collisionless, unmagnetized hydrogen plasma (m e (i) and T e (i) are the electron (ion) mass and temperature, n 0 is the unperturbed plasma density away from the grain). The dust grain charges by collecting plasma and emitting electrons. The dynamics of the system...