The widespread utility of lasers throughout science and technology is largely due to their ability to produce a beam of light with an exceedingly large degree of optical coherence. The amount of optical coherence within a continuous-wave laser beam can be quantified by the dimensionless number C, defined as the mean number of photons in the most populated mode of the beam. By analysing this quantity for “standard” laser models as found in numerous quantum optics textbooks, one finds the standard quantum limit C ∼ 8μ2 in an ideal regime, where imperfections such as technical noise are negligible. Here, μ represents the total number of coherent excitations stored within the laser device, and is the energy resource that is used for the production of coherence within the beam. However, recent theoretical work has demonstrated that there can be a quantum enhancement for the production of C. Without making conventional assumptions about how a laser operates, it was shown in Ref. [BSBW20] that C = Θ(μ4) represents an achievable upper bound—i.e., the Heisenberg limit—for the production of coherence by a laser device. The existence of this dramatic quantum enhancement in the production of optical coherence could have far-reaching implications within the established fields of quantum optics and laser physics. Broadly speaking, this two-part thesis pursues a number of research avenues that have been opened by this recent development. The first part of this thesis is concerned with generalising the theory surrounding the Heisenberg limit for laser coherence. Specifically, the achievable upper bound C = Θ(μ4) is made more robust, such that it applies to a much broader set of beams compared to that in the original work of Ref. [BSBW20]. Included within this set of beams are those that exhibit a significant degree of nonclassicality, quantified by the occurrence of sub-Poissonian photon number fluctuations. By developing a number of families of laser models that can produce beams with both Heisenberg-limited coherence (where C saturates the μ4 upperbound scaling law) and a significant degree of sub-Poissonianity, we find the absence of a tradeoff between these two quantities. That is, by modifying the system parameters to increase the degree of sub-Poissonianity, we also see an increase in the coherence. Through a detailed analysis of these families of laser models, we obtain insight into this curious phenomenon, as well as general features of Heisenberg-limited lasers that highlight their similarities with and differences from standard laser models, which instead achieve C ∼ 8μ2 at best.

The second part of this thesis is more directly concerned with practical objectives. There, we investigate the feasibility of experimentally realising a device that can produce a beam with a value of C that surpasses the standard quantum limit. Inspired by recent dissipation engineering experiments in the field of circuit quantum electrodynamcis (QED), we develop several models that can realise laser-like dynamics. Among these are models that exhibit the nonlinear dissipative interactions that are required to produce a beam with an with an improved value of C, beyond the standard quantum limit. We perform a detailed numerical analysis on one of these systems, considering parameter values that are in line with what can be achieved with contemporary circuit QED technology. From this analysis, we identify key processes within the dynamics that should be mitigated when attempting to physically implement the model to achieve our objectives. This should guide experimental efforts directed toward the general goal of demonstrating an improvement in the production of optical coherence, beyond the standard quantum limit.