Experiments

To determine the optimal shape and size for the filter medium, the Engineering Team conducted an initial experiment. Volumes of 10, 20, and 30 mL, as well as a thin/wide tray shape and a tall/thick tube shape were compared. After the completion of the experiment, the team concluded that the 30 mL tray filter media performed the best because of their uniform structure. However, the tray-shaped filter media have a thin film-like layer on the surface which may inhibit mass exchange between the atmosphere and the internal filter medium. It was decided that the tray filter media were optimal as they appeared to have a more uniform structure and macropore size, and the film-like layer could be removed following the freeze-drying if needed. In terms of volume, as seen in Table 1 the 10 mL filter medium in trays was very thin, causing it to be easily broken. Although the 20 mL filter medium was structurally stronger than the 10 mL media, the 30 mL medium was chosen because a thicker filter will provide more nutrients and space for the Saccharomyces cerevisiae cells to grow on and attach to. Furthermore, the 30 mL filter medium appeared more porous with visual signs of relatively uniform macropores.
Table 1

Filter Medium Shape	Filter Medium Volume	Observations	Pictures
Tray	10 mL	Film-like layer on surface, relatively uniform properties.
	20 mL	Film-like layer on surface, relatively uniform properties.
	30 mL	Film-like layer on surface, most porous out of the trays, relatively uniform properties.
Tube	10 mL	Malleable (bounces back when compressed), relatively uniform properties, irregular shape.
	15 mL	Very porous near the tip, more compact on top surface, irregular shape.

To achieve a baseline understanding of yeast growth on YTG agar in comparison to YTG agarose, a serial dilution of yeast broth culture from 10^-1 to 10^-6 was spread plated on both media, with colony forming units (CFUs) presented in Table 2.
Table 2: CFU counts of S. cerevisiae on 1.5% YTG agar and 5% YTG agarose media across serial dilutions. Yeast cultures were plated on YTG agar and YTG agarose plates using dilutions 10⁻¹ to 10⁻⁵. CFU counts at higher dilutions of 10⁻⁴ and 10⁻⁵ are reported. Colonies were TNTC at 10⁻¹ to 10⁻³.

Dilution	Agar	Agarose
10-1	TNTC	TNTC
10-2	TNTC	TNTC
10-3	TNTC	TNTC
10-4	209	180
10-5	25	24
10-6	4	5

On both YTG agar and YTG agarose, at the lower dilutions of 10⁻¹ to 10⁻³ colony growth was TNTC (too numerous to count), indicating high initial cell density. At the 10^-4 dilution, CFU counts were 209 on YTG agar and 180 on YTG agarose, showing relatively comparable colony counts, with slightly more on agar.
Overall, the results suggest that yeast viability is similar on both 1.5% YTG agar and 5% YTG agarose. Minor differences in colony counts were observed, but these were not substantial, suggesting that YTG agarose may serve as a suitable alternative to traditional agar. Further trials are recommended to confirm reproducibility, as well as the exploration of various other YTG agar and YTG agarose concentrations to expand understanding of yeast growth.

To assess the growth of Saccharomyces cerevisiae when submerged in water and to determine if S. cerevisiae prefers to adhere to the filter media or suspend in water, the team conducted spread plate experiments using broth culture diluted to 10⁻4 and 10⁻⁵. After spread plating 100 µL of diluted broth culture, two conditions were tested. The first was to immediately cover the plate with 20 mL of water, while the other was to incubate the plate for 2 days to let the yeast grow before covering with 20 mL of water. Through this experiment, the team hoped to determine if allowing the yeast to establish itself on the filter medium would encourage it to adhere to the filter media instead of suspending in the water.
Samples were taken from each plate for 2 days after the addition of water and were analyzed for optical density (OD) and colony forming units (CFUs), which are both able to provide information on yeast that may transfer into the water from the filter medium. The OD measures the cloudiness of a liquid and was used to quickly estimate the relative amount of yeast in the water. On the other hand, the CFU plate counts take more time but give a more accurate estimate of the amount of yeast in the water. The results of this experiment are presented in Table 3 and 4.
Table 3: CFU counts and OD_600nm readings in MilliQ water covering plate (Condition 1). MilliQ water samples were plated on standard YTG agar plates. Samples were taken from plates that were immediately covered with 20 mL of milliQ after inoculation with 10^-4 and 10^-5 dilutions of yeast culture.

Dilution	OD of water in plate		CFU count from water in plate
	Day 1	Day 2	Day 1	Day 2
10-4	0.014	0.038	0	80
10-5	0.015	0.032	0	1

Table 4. CFU counts and OD_600nm readings in MilliQ water covering plate (Condition 2). MilliQ water samples were plated on standard YTG agar plates. Samples were taken from plates that were incubated for 2 days after inoculation with 10^-4 and 10^-5 dilutions of yeast culture, then covered with 20 mL of MilliQ.

Dilution	CFU count after 2 days incubation	OD of water in plate		CFU count from water in plate
		Day 1	Day 2	Day 1	Day 2
10-4	180	0.423	0.529	TNTC	TNTC
10-5	27	0.746	0.821	TNTC	TNTC

The general trend of the data shows that S. cerevisiae prefers to suspend in liquid rather than adhere to the filter medium. This behavior was shown best with the plates that were incubated for 2 days after inoculation then covered with 20 mL MilliQ water. Shown in Table 4, there is a significantly higher amount of yeast in the water. While this is an indication that the yeast is more easily washed off after establishing itself on the filter medium in condition 2, it is possible that there is more yeast present in general, due to better growth conditions near the beginning, compared to condition 1 plates that were immediately covered in water. The preference S. cerevisiae has for dispersing in liquid is also seen after day 1 of adding water in the condition 2 plates (Figure 1), as the colonies were found to expand and any movement caused them to be swept away.

Whereas the results of this experiment give some insight into the ability of S. cerevisiae to adhere to the filter medium, the results are inconclusive as there are too many variables to consider. Additional experiments must be conducted for more concrete results.

To investigate whether the agarose filter medium behaves differently from standard agar media due to compositional differences, a serial dilution of yeast broth culture (10⁻¹ to 10⁻⁵) was plated in duplicate on YTG agar and YTG agarose plates with the same concentrations of agar or agarose (1.5% w/v). Two independent trials were conducted to ensure reliability and reproducibility, presented in Table 5 (Trial 1) and Table 6 (Trial 2). In the first trial, dilutions of 10⁻¹ to 10⁻³ consistently resulted in colony forming units (CFUs) that were too numerous to count (TNTC) and were thus omitted from the second trial.
Table 5. CFU counts of S. cerevisiae on 1.5% YTG agar and 1.5% YTG agarose media across serial dilutions (Trial 1). Yeast cultures were plated on agar and agarose plates in duplicate using dilutions 10⁻¹ to 10⁻⁵. CFU counts at higher dilutions of 10⁻⁴ and 10⁻⁵ are reported. Colonies were TNTC at 10⁻¹ to 10⁻³.

Dilution	Agar (series 1)	Agar (series 2)	Agarose (series 1)	Agarose (series 2)
10-1	TNTC	TNTC	TNTC	TNTC
10-2	TNTC	TNTC	TNTC	TNTC
10-3	TNTC	TNTC	TNTC	TNTC
10-4	248	271	202	303
10-5	9	31	22	31

Table 6. CFU counts of S. cerevisiae on 1.5% YTG agar and 1.5% YTG agarose media across serial dilutions (Trial 2). Yeast cultures were plated on YTG agar and YTG agarose plates in duplicate using dilutions 10⁻³ to 10⁻⁵. CFU counts at higher dilutions of 10⁻⁴ and 10⁻⁵ are reported. Colonies were TNTC at 10⁻³.

Dilution	Agar (series 1)	Agar (series 2)	Agarose (series 1)	Agarose (series 2)
10-3	TNTC	TNTC	TNTC	TNTC
10-4	288	142	261	189
10-5	44	12	24	32

At lower dilutions (10⁻⁴ and 10⁻⁵), colonies were countable and allowed for comparison between YTG agar and YTG agarose. Overall, both media types supported robust yeast growth, with no significant differences in CFU counts between agar and agarose at the same dilutions. This trend was consistent across both trials, indicating that agarose can effectively support yeast proliferation in solid culture. CFUs were counted after incubation, and differences in colony morphology were noted. Although CFU numbers were generally comparable between media types, colonies on agarose were consistently smaller in size than those on agar (FIGURE 2), potentially due to differences in nutrient availability or matrix structures.
Thus, these results suggest that while both agar and agarose media support similar levels of yeast growth in terms of CFUs, their physical properties may influence colony morphology differently. Media with varying agar concentrations and mixtures of agar and agarose should be further investigated to determine the best media composition for yeast growth.

To evaluate the effect of different solid media compositions on yeast growth, Saccharomyces cerevisiae cultures were plated on media containing varying agar concentrations (1.5%, 3%, 5%) as well as a 50:50 agar/agarose mixture. Serial dilutions (10⁻⁴ and 10⁻⁵) were plated on each medium type, and colony-forming units (CFUs) were counted after incubation (Table 7).
Table 7. CFU counts of S. cerevisiae on 1.5% YTG agar, 3% YTG agar, 5% YTG agarose and a 50:50 YTG agar/agarose mixture. east cultures were plated on YTG agar and YTG agarose plates in duplicate using dilutions 10⁻⁴ to 10⁻⁵.

Dilution	1.5% agar	3% agar	5% agar	50/50 agar/agarose
10-4	102	378	TNTC	TNTC
10-5	34	36	42	44

Similar growth was observed across all media types except the 1.5% agar, which had the least CFUs. These results suggest that higher concentrations, either through increasing agar content or by incorporating agarose, will enhance yeast growth on two-dimensional media. While the 3% agar, 5% agar, and the 50:50 agar/agarose plates performed similarly, the limited number of replicates makes it difficult to determine whether these media are functionally equivalent. Additional trials with replicates are recommended to assess subtle differences in yeast growth between these conditions more accurately.

To evaluate the feasibility of producing porous agar-based filter media via freeze-drying, agar media of varying concentrations (1.5%, 3%, and 5% w/v) were prepared and subjected to the freeze-drying process (Figure 3). Filter media structural integrity, texture, and porosity were assessed qualitatively post-drying.

Filter media made from 1.5% agar (Figure 3a) were weak and highly brittle. When removed from the molds, both water-rehydrated and broth-rehydrated filter media broke into layered fragments, suggesting internal weaknesses resulting from either phase separation or freezing during the drying process.
The 3% agar (Figure 3b) filter media also lacked structural integrity. Although they retained their general shape, both filter media were soft and spongy, releasing a gel-like substance when compressed. This is potentially due to residual moisture and incomplete freeze-drying.
In contrast, the 5% agar (Figure 3c) filter media demonstrated relatively strong structural integrity. Both filter media maintained their shape, showed uniform pore structure, and were not excessively brittle. They could be handled without crumbling and displayed the most favorable characteristics among the concentrations tested. However, the properties of the freeze-dried agarose media were best. The agarose media was selected for use in the filter medium because of its better properties when freeze dried and ability to support comparable growth to agar.

As a follow-up for the Yeast Growth Underwater experiment, 100 µL of pure broth culture was spread plated on several medium compositions to determine the extent of S. cerevisiae adhesion to different filter media. Yeast adhesion was assessed by swabbing plates with established yeast growth before and after washing with water, incubating the swabs in broth overnight, then measuring the OD_600nm of the broth samples (Table 11). The differences between OD measurements before and after were then calculated to determine the extent of yeast lost in the water.
Table 11. S. cerevisiae lost to washing 2 and 9 days after plate inoculation measured by OD_600nm of swabs. Swabs were taken by twisting a cotton swab on the surface of the plate. Washes occurred by pipetting 1 mL of sterile MilliQ water onto a portion of the plate with the swabbing occurring over the impacted spot.

Iteration	Nutrients	Medium	Zeolite	Pre-Wash		Post-Wash		Difference
				Day 2	Day 9	Day 2	Day 9	Day 2	Day 9
A	1:5 dilution	Agarose	1% (w/v)	0.842	0.947	0.008	0.267	-0.834	-0.680
A	1x			0.727	0.001	0.644	0.006	-0.083	0.005
B	1x			1.345	0.041	0.733	0.002	-0.612	-0.043
A	2x			0.288	1.555	0.385	0.686	0.097	-0.869
B	2x			0.733	-	0.423	-	-0.310	-
A	1:5 dilution		0.5% (w/v)	0.128	0.217	0.006	0.083	-0.122	-0.134
A	1:5 dilution			0.621	0.263	0.006	0.793	-0.615	0.530
B	1x			0.392	1.073	0.11	1.507	-0.282	0.434
A	1x			1.507	1.071	0.11	1.507	-1.397	0.436
B	2x			1.373	1.471	0.084	0.987	-1.289	-0.484
A	1:5 dilution	0%		0.092	1.261	0.005	0.083	-0.087	-1.178
B	1:5 dilution			0.985	0.789	0.426	0.474	-0.559	-0.315
A	1x			1.165	1.05	0.634	0.606	-0.531	-0.444
B	1x			0.678	0.801	0.163	0.286	-0.515	-0.515
A	2x			0.029	0.435	0.062	0.637	0.033	0.202
B	1x			1.403	1.553	0.002	0.001	-1.401	-1.552
Agar							1.293		0.243
Average				0.806	0.808	0.193	0.322	-0.578	-0.468

The results (Table 11) show that the incubation period of the filter medium is insignificant. Compared to the 2-day OD readings, any increases in the 9-day readings are generally negligible, with the average OD readings both around 0.8. Similarly, the difference in OD from the 2- and 9-day readings appear to be relatively comparable. In terms of before and after washing, the average OD difference was around 0.5. When compared to the average initial OD of 0.8, it shows that the S. cerevisiae does not adhere well to the filter medium, with over half of the yeast being washed away by the water.
Based on overall reproducibility, extent of yeast lost in the water, and overall yeast growth, the 1x nutrient concentration with agarose and 1% zeolite is the optimal choice of filter medium. This composition resulted in the lowest OD_600nm difference of 0.005 and 0.043 on day 9 and had a relatively low OD difference of 0.083 and 0.612 on day 2. Furthermore, the average OD reading for pre-wash yeast growth was 1.03, which is above average. Other filter media compositions that may be considered due to reproducibility, yeast loss in water, and overall yeast growth include 2x nutrient concentration with agarose and 1% zeolite, as well as 0.5% zeolite.
Whereas the results of this experiment provide more reliable information about the ability of S. cerevisiae to adhere to the filter medium, there was generally high variability between iterations of the same conditions. Thia variability is likely due to human error during swabbing. Thus, this experiment should be performed with more replicates to verify the results.

To determine the ability of water to permeate through the filter medium, 100 µL of dyed water was deposited on top of a small sample of the medium. Trials were conducted in duplicate with the film-like layer on the surface removed and kept. When water was deposited on the sample with the film-like layer still intact the water slid off the medium, indicating that the top layer of the filter medium is significantly less permeable than the interior and must be removed for filtration applications. The results of this experiment for the sample with this layer removed (Table 12) prove that water can permeate through the filter medium. Table 12. Water permeation into the filter medium, measured by depositing dyed water on top of a sample.

Time	Distance Permeated (mm)		Instantaneous Permeation Rate (mm/h)
	Iteration 1	Iteration 2	Iteration 1	Iteration 2
10 min	6	8	36	48
30 min	7	10	14	20
1 h	7	10	7	10
2 h	10	12	5	6
5 h	11	13	2.2	2.6
15.5 h	15	15	0.97	0.97
20.5 h	18	16	0.88	0.78
48 h	22	21	0.46	0.44
Average			8.31	11.10

Under the influence of gravity, the water is shown to very slowly permeate through the filter medium, with an average flow rate of 8.31 mm/h and 11.10 mm/h for different iterations with the same conditions. Applying pressure to the inflow of water increases the rate of permeation through the medium, which was tested by forcing a faster inlet flow rate using a dropper. This indicates that the rate of water filtration through the medium can be easily controlled by altering the flow rate of water into the medium.

An inoculated filter medium was cut to expose its interior cross section, which was analyzed using phase contrast microscopy. The presence of many yeast cells on the surface of the cross section proves that yeast cells can move through the porous structure and colonize the filter medium because the inoculant was added to the filter medium's exterior. Thus, the inoculant cells must have moved through the filter medium to be present on the interior cross-section sample imaged. The presence of yeast cells throughout the pore is evident by zooming into the pore and observing cells at different depths. The imaged pore is a close pore. Therefore, this experiment confirmed that the pore structure is comprised of interconnected and sealed pores and that yeast can move through and colonize the pores.