Development of efficient, selective and cost-effective electrocatalysts is central to the efficient production of sustainable fuels and industrial feedstocks. While significant progress has been made in the hydrogen evolution reaction (HER) for water splitting, the direct synthesis of industrial fuels and feedstocks by the CO2 reduction reaction (CO2RR) still presents significant challenges related to the limited number of metals that maintain selectivity toward multi carbon products while operating at a low overpotential.Recent experimental work by Choi and co-workers1 has demonstrated the generation of 1-butanol from the reduction of CO2 on phosphorus-rich copper cathode. 1-butanol was produced at a record faradaic efficiency of approximately 4%. Interestingly, it was observed that carbon monoxide, CO, which is considered the preeminent intermediate for CO2 reduction reaction was not observed as an intermediate in this reaction. On previous copper based electrocatalysts, it has been shown that the C-C coupling through *CO dimerization is the rate limiting step that controls the selectivity toward multi carbon products.2-4 The absence of *CO as an intermediary for 1-butanol formation on CuP2 suggests the existence of a different mechanism is at a play.This presentation reports the results of computer simulations of this system based on Joint Density Functional Theory, JDFT,5 which combines a first-principles treatment of the electrocatalyst (CuP2) and adsorbed molecular fragments together with a realistic continuum model of other parts of the electrochemical environment. The simulations are used to explain the potential dependence and the mechanism of 1-butanol formation, and other experimentally observed products, on a CuP2 electrocatalyst. Traditional theoretical studies on electrocatalyst have used high vacuum DFT simulations, by using the computational hydrogen electrode, CHE, model.6 CHE presents a fundamental limitation on the description of electrochemical systems because it largely ignores solvation and local electric field effects, which are vital for the accurate description of electrochemical systems. Explicit inclusion of water into the theoretical simulation isn’t only limited by the many solvent molecules needed but also thermodynamic sampling of many configurations of those molecules to properly capture the structure and response of the liquid over realistic experimental time scales and lengths. In response to the above limitations, continuum solvation models are employed, in which the details of the molecular aqueous environment are replaced by a continuum description. In addition to solvation effects, the effect of the applied potential on the reaction is also investigated. The applied potential is represented by employing a grand canonical ensemble.Using the JDFT formalism, many configurations of molecular fragments on the CuP2 catalyst were considered and their optimal structures and energies were obtained. It is shown that the phosphorus atoms on CuP2 act as hydride binding sites. Thermodynamic and kinetic results show that the activation of physically adsorbed CO2 by chemically adsorbed H favors the formation of formate (HCOO) which is adsorbed on Cu sites through its oxygens, over the formation of carboxyl (*COOH) intermediate. The unfavorable carboxyl intermediate likely explains why *CO is not observed as an intermediate in experimental work.1 The formation of multi-carbon products is controlled by the C-C coupling step. We show that the coupling step occurs through the aldol condensation first between two formaldehyde then through two acetaldehyde adsorbates. These findings not only explain the experimentally observed products but also illustrate an alternative mechanism toward multi-carbon formation on electrocatalysts that does not involve CO-CO coupling step.Reference M. Choi, S. Bong, J. Won Kim, J. Lee. ACS. Energy. Lett. 2021 , 6 , 2090-2095J. Montoya, C. Shi, K. Chan, J. K. Norskov. J. Phys. Chem. Lett. 2015, 6, 2032-2037J. Montoya, A. Peterson, J. K. Norskov. Chem. Cat. Chem. 2013, 5,737-742H. Xiao, T. Cheng, W. A. Goddard, R. Sundararaman. J. Am. Chem. Soc. 2016, 138, 483-486R. Sundararaman, K. Letchworth-Weaver, K. A. Schwarz, D. Gunceler, Y. Ozhabes, and T. A. Aria, SoftwareX, 2017, 6, 278-284; K. Schwarz and R. Sundararaman, Surface Science Reports, 2020, 75, 100492A. Peterson, F. Abild-Pedersen, F. Studt, J. Rossmeisl, J. K. Norskov. Energy Environ. Sci. 2010, 3, 1311-1315 Acknowledgement:This work was supported by the Wake Forest Chemistry Department. Computations were performed on the Wake Forest University DEAC Cluster, a centrally managed resource with support provided in part by the University.