This thesis is positioned within the field of green logistics with respect to CO2 emissions in road freight transportation. In order to examine the different aspects of CO2 emissions of freight transportation, three related, but different research questions are studied. Because CO2 emissions are proportional to the amount of the fuel consumed by vehicles, the first goal of the thesis is to review and compare several available fuel emission models. The results of extensive computational experiments show that all emission models tested are sensitive to changes in load, speed and acceleration. Second, the dissertation studies the Pollution-Routing Problem (PRP), an extension of the classical Vehicle Routing Problem with Time Windows (VRPTW). The PRP consists of routing a number of vehicles to serve a set of customers within preset time windows, and determining their speed on each route segment, so as to minimise a function comprising fuel, emission and driver costs. A mathematical formulation of this problem cannot be solved to optimality for medium to large scale instances. For this reason, the thesis describes an adaptive large neighbourhood search (ALNS) based algorithm to solve the PRP. The algorithm iterates between a VRPTW and a speed optimisation problem, where the former is solved through an enhanced ALNS and the latter is solved using a polynomial time speed optimisation algorithm (SOA). The third question relates to the PRP and the two important objectives that should be taken into account, namely minimisation of fuel consumption and total driving time. Computational results on a large set of PRP instances show that the algorithm is both effective and efficient in solving instances of up to 200 nodes. The thesis therefore studies the bi-objective PRP where one of the objectives is related to the environment, namely fuel consumption (hence CO2 emissions), and the other to driving time. An enhanced ALNS algorithm is described to solve the bi-objective PRP. The algorithm integrates the classical ALNS scheme with a specialized SOA. The results show that one need not compromise greatly in terms of driving time in order to achieve a significant reduction in fuel consumption and CO2 emissions.