Open Collections

Investigation into the factors affecting controlled struvite crystallization at the bench-scale

University of British Columbia

This research was initiated with the purpose of understanding more about the process of struvite crystallization - a novel way of recovering phosphorus as phosphate from wastewater. A new reactor design was tested at the bench-scale, in terms of its ability to provide extra turbulence inside the reactor, in order to enhance the mechanical strength of the growing crystals and by that function, make for a more convenient harvesting operation. Selective withdrawal of hard crystals was possible after minor modifications to the original design, which was eventually adopted at the pilot-scale as well. An effort was made to arrive at some understanding, with respect to the use of the recycle stream as a tool to maintain the strength of feedwater in the reactor at some optimum concentration. Accordingly, three sets of runs, each with a different feedwater concentration (influent phosphate values of 40 mg •L-1 and 80 mg •L-1) and recycle ratio were conducted. Although the hypothesis justifying the use of a recycle stream as a diluent for the incoming phosphate stream seemed logical, the data collected from the three sets could not corroborate it. The restriction in the value of maximum allowable operational pH (pHlim) was identified as the reason for this shortcoming. An increase in the operational pH beyond pHlim was shown to be beneficial for P-removal and partly vindicated the logic for the requirement of a recycle stream. However, it was not possible to operate the bench-scale system at pH values exceeding pHlim without plugging the reactor within 24 hours after a run start-up. Limited data from the follow-up pilot-scale reactor established that it was possible to overcome the restriction of pHlim inherent at the bench-scale. The higher degree of turbulence inside the pilot-scale reactor and the increase in crystal loading over time (30 days - 40 days after run start-up), were identified as the factors affecting the limit in the operational pH at the bench-scale. Once the restriction afforded by pHlim was overcome, the recycle stream could function successfully as intended - as a diluent. A recycle stream of 6:1 and an operational pH of 8.2 were identified as the prime requirements to achieve phosphorus removals of 80%- 95% for an incoming phosphate concentration of 60mg•L-1 - 70 mg • L-1, at pilot-scale. Runs with low influent phosphate concentrations (20 mg • L-1 - 30 mg • L-1) were observed to exhibit signs of crystal growth at a relatively high pH value of 9.0, at bench-scale. This pointed to the possibility that P-recovery through crystallization from such concentrations, may not be economically feasible. The possibility of using struvite solubility criteria as a control parameter for a struvite crystallization experiment was also explored. A conditional solubility curve for struvite, generated from experiments conducted at the University of British Columbia (UBC), was used as a reference point for this purpose. However, a plot of the conditional solubility product values of the bench-scale runs against the curve, incorrectly implied that the conditions inside the reactor were undersaturated. It was hypothesized that this curve may not serve as a universal quantification of struvite solubility criteria for all struvite crystallization reactors, since the underlying reaction is highly dependent on the physical characteristics existing inside the reactor at any given time; this would include the degree of turbulence and the seed / crystal loading.

Thesis/Dissertation

application/pdf

2009-08-12

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/12074

Civil Engineering

University of British Columbia

UBCV

DSpace