A transition in the energy mix towards clean energy sources is essential to ensure environmental improvement and sustainability. This study investigates the potential impacts of nuclear energy, renewable energy and energy technologies research and development (R&D) expenditures on the carbon footprint of the United States (USA), which is a leading country in the use of nuclear and renewable energy. In the study covering the period 1974-2023, autoregressive distributed lag (ARDL) and non-linear autoregressive distributed lag (NARDL) approaches are used in order to take into account the symmetric and asymmetric effects between the variables together. Empirical findings indicate that there is a symmetric and asymmetric cointegration relationship between the variables in the long run. ARDL analysis findings show that nuclear energy, renewable energy and energy technology R&D expenditures have a significant and negative impact on the carbon footprint in the long run. The findings of the NARDL analysis, which takes into account the positive and negative shocks in the variables, indicate that, except for the positive shocks in energy technology R&D expenditures, positive and negative shocks in all variables reduce the carbon footprint in the long run. Therefore, the findings reveal that the consumption of nuclear and renewable energy significantly improves environmental quality and renewable energy is relatively more effective in reducing carbon footprint compared to nuclear energy in the country. These findings offer some important implications for the country’s energy policy.