Process Flow Diagram Fig. 1.

Process Flow Diagram Fig. 1

Figure 1. This is a flow diagram of the continuous flow ozonation apparatus used in the Department Of Energy project; ENERGY AND PROTEIN PRODUCTION FROM PULP MILL WASTES. As is shown at left, the Welshbach Ozone Generator requires very dry, pure oxygen feed and a water cooling loop to remove heat generated by internal UV radiation. It produces a continuous 98:2% oxygen to ozone ratio stream, fed into the bottom of the reactor. Desugared Spent Sulfite Liquor (SSL) is continuously introduced into the top of the reactor and removed at the bottom. The column is packed with 1.25 cm berl saddles to gain surface area and residence time for the ozone to fully react with the SSL. There is a continuous "side loop" into a small stirred flask with pH probe and automatic titrator for pH control in the column. Temperature control in the reactor was not required. Oxygen passing out the top of the reactor is bubbled through a KI absorber to trap any residual ozone prior to being vented to a hood (38, 77, 137, 195).

Go back to Selection Menu
Home Page

Table 1. Average Values for COD, BOD, Total Sulfur, and Organic Acids in Untreated and Treated SSL1

Unozonated SSL
Ozonated SSL
Fermented Ozonated SSL
70,699 mg/L
57,944 mg/L
46,712 mg/L
6,469 mg/L
7,966 mg/L
5,074 mg/L
4.59 mg/L
4.26 mg/L
3.84 mg/L
Organic Acids2
3 mM
2 mM
2 mM
8 mM
66 mM
4 mM
4 mM
4 mM
1. All values are based on a 75% concentration; 75:25 SSL to distilled water
2. Values determined from an average of 5 replicates
nd - Less than 1 mM concentrations not detectable with procedures used

Effects of Ozonation on SSL - Chemical Oxygen Demand (COD) was lowered about 18% during ozonation indicating at least a partial oxidation and breakdown of the lignosulfonate molecule (Table 1). An increase of BOD values was evidence of a production of smaller metabolizable molecular fragment from the lignosulfonate molecule. Total sulfur content was lowered about 7%, caused by a formation of calcium sulfate, which precipitated out from the liquor.

A gas chromatographic analysis of ozonated SSL indicated that significant amounts of acetic, formic, and oxalic acid were produced by ozonation (Table 1). The production of these organic acids would account for some of the BOD increase.

Fermentation Effluent Analysis - A cumulative drop of 34% was observed in COD values for a combined treatment of both ozonation and fermentation (Table 1). Fermentation of ozonated SSL reduced the average COD value of the effluent by 19%, while the average BOD value dropped 36%. There was an initial increase in BOD of about 23% after ozonation, however these organics were used up during fermentation. When unozonated SSL was fermented under similar conditions, COD values dropped only 7% with no noticeable change in the BOD value. A 16% loss in total sulfur was observed after the combined ozonation - fermentation of SSL. The 9% loss during anaerobic digestion was likely due to H2S production, however, the chromatographic column used did not detect H2S.

Organic acid analysis of SSL indicated that 66 mM of acetic acid were present in the fermenter effluent along with small amounts of propionic, formic, and butyric acids (Table 1). Most of the oxalic acid produced during ozonation was removed during fermentation.

Gas production from ozonated SSL over a 3-month period averaged 12 ml/hour from each of the three steady-state fermenters maintained at a 3.3 day retention time (Table 2). This retention time and gas production rate for an average fermenter volume of 720 ml would yield 1.76 liters of gas produced per liter of ozonated SSL. This contrasts with 2 ml/hour or 340 ml/liter from unozonated SSL. When the retention time was brought below 3.3 days, gas production rates increased 10% initially, then rapidly dropped off, implying a washing out of the gas producing bacteria. When the retention time was increased to greater than 3.3 days, a decrease in gas production was observed proportional to the increase in retention time.

The ratio of CH4:CO2 in the effluent gas was approximately 3:1 (Table 2). This is a much higher quality of gas than the 1:1 ratios reported by others to be produced from anaerobic digestion of domestic wastes, and could be due to the low carbohydrate content of SSL or to a shorter retention time than used in other anaerobic systems (38). However, the effect of an extended retention time of 12 days was tested with two fermenters fed 75% ozonated SSL continuously for 1 month. There was virtually no change in gas composition or acetic acid concentration during this period.

Table 2. Gas Production During Continuous Fermentation of Ozonated and Unozonated SSL1.

Retention Time, Days
Gas Produced, ml/hr
Carbon Dioxide
Ozonated SSL
78 %
22 %
Ozonated SSL
Wash Out2
52 %
48 %
Unozonated SSL
14 %
86 %
1. All values are based on a 75% concentration; 75:25 SSL to distilled water
2. When the retention time was brought below 3.3 days, the gas production rates
increased 10 % then rapidly dropped off, implying a wash out of bacteria.

Table 2. Is a summary of gas production during the continuous fermentation of spent sulfite liquor for the Department Of Energy project report; ENERGY AND PROTEIN PRODUCTION FROM PULP MILL WASTES.

Results and Discussion - Of the total COD decrease during fermentation, only 25% can be accounted for as the COD of methane produced. The CO2 produced during fermentation does not contribute to COD loss since it is already in the most oxidized state. The major portion of the unaccounted COD loss (8,200 mg O2/liter SSL) is probably due to bacterial oxidation reactions in which the oxygen of water is one of the reactants (38).

Some of the COD change is also due to the loss of hydrogen and sulfur as H2S. Both H2S and hydrogen in the product gas stream were below detection levels and, as such, the degree of COD loss due to these factors should be minimal. However, even on the assumption that the total sulfur loss (0.42 g/liter) during fermentation is converted to H2S, this factor could account for only 5% of the decrease in COD. It is more likely that dissolved sulfur loss was due to precipitation of FeS.

Since the reduction in BOD after fermentation varies so much with the substrate being fermented, it is difficult to make meaningful comparisons. However, an overall decrease of BOD during fermentation is within the ranges reported in the literature for domestic wastes. (38)

The average number on methanogens present in ozone-treated SSL fermentions is in the upper range of 106 - 109 methanogens/ml commonly encountered in the digestion of other organic wastes. The high population of methanogens observed in ozonated SSL fermentations as compared to sewage sludge may be due to less competition from other anaerobes in the low carbohydrate SSL or to differences in counting media and procedure (38). Sulfur-reducing bacterial populations at 1.7 X 109 counts/ml in fermenting SSL were also much greater than the 3 to 5 X 104 counts/ml commonly found in sewage digesters (38). The high sulfur content of SSL likely accounts for these higher numbers of sulfur-reducing bacteria.

Since methane bacteria have been shown to use acetate as a substrate for methane production, it is surprising to find substantial amounts of acetate(66 mM) present in ozonated SSL fermenter effluent. The net production of 58 mM (3,480 mg/liter) acetic acid during fermentations is much higher than the 200 mg/liter - 860 mg/liter range commonly reported by other investigators working with anaerobic digestion of organic wastes (38). If all the 66 mM acetate in the effluent of ozonated anaerobically digested SSL were converted to CH4, an extra 1.5 liters of CH4 could have been produced per liter of 75% SSL, a >100% increase in methane production.

A possible reason for the limited CH4 production could be the relatively high salt concentration of 13 g/liter in the 75% SSL. This is probably why SSL must be diluted to 75% with water to facilitate the growth of methanogens in this situation. Of these total salts, 7 - 10 g/liter are calcium. This concentration of is not intolerable to organisms such as the Candida utilis used to economically produce yeast protein from the residual sugars in sulfite liquor for years. However, others have concluded that a calcium level of 8 g/liter gave a definite inhibitory effect on methane production from sewage sludge (38).

Because efforts to increase methane production from ozonated SSL by the addition of growth supplements or varying fermenter operating conditions failed to isolate any positively effective variable, the effects of other pre-treatments of SSL prior to fermentation were examined. Research previously done by other groups on sewage sludge indicated that sonication greatly increased its biodegradability. A small reactor was constructed with a sonic "wafer" installed in the base and a stainless steel sparging ring positioned 0.5 cm above this wafer for injecting Ozone into the raw SSL. This reactor was operated in batch mode only. A frequency of 41,000 Hz was used in experiments with this reactor. Control experiments with Ozonation of SSL alone were performed to establish a base line for COD and BOD. Ozonation in conjunction with sonication of SSL showed a 25% increase in BOD and a decrease of 10% of COD beyond that of ozonation alone. Additional experimentation in this area was not possible due to funding and time limitations (38, 77, 137, 194, 231).

It was found that an alternative to anaerobic fermentation of Ozonated SSL for Methane production, could be aerobic fermentation for production of additional yeast protein. Experiments showed that if the Torula sp. yeast originally used at the pulp mill to remove the wood sugars, producing yeast protein for sale, was reintroduced into ozonated SSL, an additional 5.0 g/liter of yeast protein would be produced for sale (38, 77, 137, 194, 195, 231).

These studies have shown that appreciable amounts of CH4 can be produced from ozonated SSL, but are far lower than yields obtained from other organic substrates. Most of the CH4 produced in the fermentation of ozone-treated SSL originates from compounds produced during ozonation and is not a result of wood sugars and other low molecular weight compounds remaining after previous yeast fermentation. In contrast, unozonated SSL did not supply the necessary substrate for comparable CH4 production.

Although this process was not economically competitive at the time of investigation (years 1976 - 1979), with process improvements, this combined ozonation - fermentation treatment of SSL could:

  • Provide a supplemental energy source to the pulping process
  • Be used to produce additional yeast protein
  • Aid in reduction of pollutants attributed to SSL

Go back to Selection Menu
Home Page

Craig Bremmon
CEBTech Services
503 11 th Ave. SE, Suite 5
Aberdeen, SD 57401
Phone: 605 226-3595
Mobile: 605 377-4062
Fax: 605 226-3595