UCLA

Condensation is a very complicated subject to fully unveil as it constitutes a complex interplay of momentum, heat and species transport and interfacial physics. Drop-wise condensation (DWC) adds more layers of complexity by introducing droplet dynamics and the two-way interaction with droplets surrounding. DWC has been shown repeatedly for about a century to possess around an order of magnitude improved heat transfer rates. Additionally, it has been shown that DWC is limited by the maximum size of droplets a surface sustains. Since this realization, promoting DWC has been greatly steered towards developing stable hydrophobic coating techniques. However, attempts have not been feasible so far especially due to the extreme conditions encountered in condensation processes. Another concern that has not been resolved adequately is the deterioration of heat and condensation rates due to the presence of non-condensable gases (NCG). This dissertation aims at understanding the process of condensation especially in the presence of NCG.

First, a numerical model of the process of vapor condensation on surfaces characterized by film-wise condensation with the presence of (NCG) is presented. State variables in both the condensate film and the diffusion layer were solved separately and the condensation interface was used to couple the two solutions. The solution of the condensate film was obtained using well-established solutions of laminar film condensation of pure vapor. In contrast to other models surveyed, this work provides a inexpensive and accurate predictions of heat and mass transfer characteristics. We validated the work against two classical condensation problems. The model was first validated against empirical correlations and experimental work, resulting in a very good agreement. We then assessed the applicability of ignoring the condensate film effect, as performed in previous models, on the condensation processes by observing the thermal resistances of both the condensate film and diffusion layer. Results indicated that for the studied cases of NCG mass fractions above 20\%, the condensate thermal resistance was at least an order of magnitude lower than that of the diffusion layer. However, the two thermal resistances seem to approach each other as NCG mass fraction becomes smaller. On another front, we observed that models that ignore the condensate film thermal resistance underestimate the interfacial temperature albeit accurately predicting the overall heat transfer rate. To simulate even lower NCG mass fractions, we validated our model to the classical analytical work of Sparrow and co-workers. Results showed an excellent agreement between the two solutions at different NCG mass fractions (0.5\%-10\%) and subcooling degrees (5$^o$F-40$^o$F). Finally, we found a good agreement between results of our model and the heat/mass transfer analogy. The heat/mass transfer analogy is a semi-empirical method therefore, is limited to the existing correlations and their uncertainties. On the other hand, our model does not use any empiricism and relies on the available solutions of laminar condensate film of pure vapor in predicting the liquid side heat transfer coefficient.

Moreover, motivated by the improvement in heat transfer by frequently disturbing the thermal boundary layer in nucleate boiling phenomenon, we attempted making the analogy to DWC. We developed an initial theoretical model to predict the transience of the diffusion boundary layer in condensation problems with NCG presence. The problem of suddenly exposing a cooled surface to a humid environment was modeled as two semi-infinite gas and solid domains in contact with the condensate film as a coupling condition. Results showed that the transient behavior of heat and condensation rates start by very high values and then decrease to steady state rates. This suggested that if the diffusion layer is frequently disturbed by droplets of heights similar to the layer's thickness, the condensation rate is expected to improve. To realize this, jet impingement of humidified air was proposed as a means of thinning the diffusion layer as well as to provide better shedding capabilities of condensate droplets.

Finally, Utilizing jet impingement technique as a means for continuous drop-wise condensation (CDC) was investigated. The technique showed an advantage of overcoming the necessity of using highly non-wetting surfaces while yet maintaining micron-sized droplets. By shifting focus from surface treatment to the force required to sweep off a droplet, we were able to utilize stagnation pressure of jet impingement to tune the shed droplet size. To demonstrate the effectiveness of this technique, we performed condensation experiments on a broad range of contact angle and contact angle hysteresis surfaces. The results showed that droplet size being shed can be tuned effectively by tuning the jet parameters namely the jet Reynolds number. Droplets as low as 20 $\mu$m in radius on a hydrophilic surface were shed with this technique surpassing the traditional gravity-assisted shedding mechanism by almost 80 folds. In terms of condensation rate improvement, we showed theoretically that CDC improves the condensation rate of pure steam, and hence heat transfer rate, by more than 300\% compared to gravity-assisted shedding DWC. Finally, our experimental observations showed that the effect of NCG, such as air in this work, is greatly alleviated by utilizing our technique. An improvement by at least six folds in mass transfer compactness factor compared to state-of-the-art dehumidification technology was possible. We illustrated the physics of droplet departure and mobility due to the stagnation flow condition by microscopically tracing a single droplet.