Research of novel process route and scale-up based on oil-water separation ﬂ otation column

A novel process ‘ coalescence-air ﬂ otation-carrier preferential adsorption ’ utilising an oil-water separation ﬂ otation column with a unique structure was used in the oil-water separation ﬁ eld. The oil-water separation ﬂ otation column contains the cyclonic separation and air ﬂ otation separation which has advantages in oily sewage treatment, especially in polymer- ﬂ ooding-drive oily sewage treatment. In this paper, different dimensions of ﬂ otation column with 1 m 3 d (cid:1) 1 , 30 m 3 d (cid:1) 1 and 2,000 m 3 d (cid:1) 1 oil-water separation systems were investigated. In addition, several operating parameters which impact separation, such as feeding speed, aeration rate, circulating pressure, adsorbents consumption and frother consumption were also investigated. The optimum operating parameters determined for 1 m 3 d (cid:1) 1 the oil-water separation ﬂ otation column were a feeding speed of 0.042 m 3 h (cid:1) 1 , an aeration rate of 0.10 m 3 h (cid:1) 1 , a coal consumption of 4 (g coal)·(g oil) (cid:1) 1 , a frother consumption of 10 mg L (cid:1) 1 , and a circulating pressure of 0.12 MPa. The novel process cost reduced 55.8% than conventional two-stage air ﬂ otation process. In the 2,000 m 3 d (cid:1) 1 oil-water separation experiment, the oil concentration and the oil removal ef ﬁ ciency of outlet are 23.39 mgL (cid:1) 1 , 97.70%, respectively. Sediment is not produced during the oily sewage treatment using the novel process and ﬂ otation column. A novel process preferential

The coalescence-airflotation-carrier preferential adsorption process has been introduced to the oil-water separation field (Li et al. ). The droplets are also coalesced when they pass through the sieve plate in flotation column. The effect is improved oil-water separation. The emulsification oil droplets are transported from the cyclone separation zone to the airflotation separation zone by the oil-gas complex which formed through the transportation role of the microbubbles, the oil-water separation has finished (Moruzzi & Reali ).
The airflotation separation area has a static separation effect of the 'long and narrow' environment with 'quiet' fluid dynamics (Xu et al. ).
The oil-water separation flotation column is integrated with the cyclonic and flotation separation technology (Lee

Experimental method
In the experiment, the oil removal efficiency was used to evaluate the oil-water separation efficiency of the flotation column. Samples were taken from the sampling point of the feeding port, the sampling point of purified underflow outlet, and the oil concentration at these three points is C 1 (oil concentration of inlet, mg L À1 ) and C 2 (oil concentration of outlet, mg L À1 ). The SS concentrations are m 1 and m 2 for the SS concentration at the inlet and outlet, respectively.
The oil removal efficiency R 1 is: The SS removal efficiency R 2 is: The oil concentration was measured using UV-4802S by ultraviolet spectrophotometer method (Zhu et al. ). Petroleum ether was used as extraction agent. The 256 nm wavelength was used in the spectrophotometer UV method. The working curve method was applied to the determination and blank solution was used as reference (Pang et al. ).
1 m 3 d À1 Oil-water separation experimental system The flotation column in this experimental system has a diameter of 100 mm, height of 2,000 mm, and is made of stainless steel.
Firstly, the oily sewage from the primary separation tank entered into the mixing tank and its flow was regulated. The The process on flotation column Step 1 -Experimental system of the coalescenceairflotation process on the flotation column The flotation column in this experimental system has a diameter of 100 mm, height of 2,000 mm, and is made of stainless steel.
The experimental system of the coalescence-airflotation process is the same as 30 m 3 d À1 oil-water separation experimental system, but the flotation column in this system is filled by coalescent pack which is shown in Figure 5. The    Table 1.
Step 2 -The experimental system of coalescenceairflotation-carrier preferential adsorption process on flotation column As shown in the results of the preliminary research, the coalescence-airflotation process is good for removing large oil droplets, but is not good for small oil droplets. The oil concentration at the outlet is in the range from 100 to 200 mg L À1 and so cannot meet the requirement for reinjection. In the carrier preferential adsorption process, coal consumption is proportional to the oil concentration at the inlet, so the coal consumption is large when the oil concentration of the inlet is high. On this basis, the proposal has arisen to combine the respective advantages of both the coalescence-airflotation process and the preferentialcarrier-adsorption process, using a combined system within the flotation column.
The combined coalescence-airflotation-preferentialcarrier-adsorption process using flotation columns is shown in the Figure 6. Firstly, the oily sewage, with or without a little frother, entered into the first mixing tank as the feeding sewage of the primary coalescence-airflotation process using the first flotation column. After flotation separation and coalescence oil removal, the oily scum (which is composed of oil, bubble, oil-bubble complex and SS) is collected by foam tank and then discharged. The effluent water of the primary coalescence-airflotation process with the sorbent (powder coal) then entered into the second mixing tank, to undergo the carrier-preferential adsorption process using the second flotation column, and after fully mixing was pumped into the flotation column.
After thorough flotation separation, coalescence oil removal and carrier-preferential adsorption, the oily scum composed of oil, powder coal, bubble, oil-bubble complex and SS is collected by foam tank and then discharged. The clean water is discharged from the bottom of the second flotation column.
The operational parameters for the process of coalescence-airflotation-carrier preferential adsorption are shown in Table 2.
2,000 m 3 d À1 Oil-water separation industrial experimental system Based on the research results, the coalescence-airflotationcarrier-preferential adsorption process was used in the 2,000 m 3 L À1 oil-water separation industrial experimental system. The 2,000 m 3 L À1 industrial system diagram of the process is shown in Figure 7. The whole industrial system includes the coalescence-airflotation separation system, the carrier preferential adsorption separation system, and an adsorbent preparation system and pressure filtration 1.25 0.20 1.5 Figure 6 | The combined process of coalescence-airflotation-carrier preferential adsorption.  Table 3.

RESULTS AND DISCUSSION
Results of 1 m 3 d À1 oil-water separation experimental

Operational parameter optimization
This paper has investigated the impact of the factors, including feeding speed, aeration rate, circulating pressure, frother consumption and powder coal consumption, on the oil/ water separation efficiency. The results are shown in   Figure 8(c), with increasing powder coal consumption, the oil removal efficiency showed a decreasing trend. With the increasing of powder consumption, the adsorption probability between the oil droplets and powder coal increases. This is advantageous for oil-water separation. Considering the requirements and cost, powder coal consumption determined by the test is 4 g coal g oil À1 .

As shown in
As shown in Figure 8(d), with increasing frother consumption, the oil removal efficiency tended to increase, but decreased after the frother consumption was more than 15 mg L À1 . After adding the frother, the bubble size reduces and the bubble quantity increases and the collision probability between bubbles, oil droplets and powder coal increases. The frother can also increase the quantity of the hydrophobic groups of the oil droplets flocs, so the adhesive quantity and quality of the bubble were enhanced and the flotation effect was improved. However, after frother consumption was more than 15 mg L À1 , this generates too many microbubbles which has a negative influence on the flow condition of the flotation column, so the separation efficiency decreases. The frother consumption determined by the test is 10 mg L À1 .
As shown in Figure 8(e), with increasing feeding speed, the oil removal efficiency showed a decreasing trend. On increasing feeding speed, the collision probability between the oil droplets and bubble reduces since the separation time of oil-water coalescence is gradually shorted. This is a disadvantage of oil-water separation. The proper feeding flow determined by the test is 0.042 m 3 h À1 .

Result of continuous operation experiment
Based on the operation parameters optimization, the 48 hours of continuous operation experiment was investigated for researching the reliabilities of the flotation column (Ø 100 mm × 2,000 mm) and the process. Water samples were taken every two hours, the water samples at the inlet and outlet were taken at the same time. The SS concen- trations of inlet samples were tested every 16 hours and were 106, 74 and 84 mg L À1 .
The circulating pressure was 0.12 MPa, the feeding rate was 0.042 m 3 h À1 , the powder coal consumption was 4 g coal g oil À1 , the frother dosage was 10 mg L À1 , and the experiment temperature was 20 W C. The results are shown in Figure 9. The results of the polymer retention effect are shown in Table 4.
As shown in Figure 9, the average oil concentration at the inlet was 1,364.14 mg L À1 , the average SS concentration at the inlet was 88.00 mg L À1 . After the oil-water separation using flotation column, the average oil concentration at the outlet was 22.24 mg L À1 and the average SS concentration at the inlet was 29.59 mg L À1 . The efficiencies of oil removal and SS removal were 98.37% and 66.38%, respectively.
According to Table 4, the powder coal cannot remove all the polymers, most of the polymers are retained in the outlet water. So the adsorption of powder coal has selectivity, good adsorption efficiency for hydrocarbons (such as oil droplets), good adsorption efficiency for polymers and SS.
According to the continuous experiment results, the process and flotation column can be used on the oily sewage treatment.

Results of 30 m 3 d À1 oil-water separation experiment
Based on the results of the 1 m 3 d À1 oil-water separation experiment, a 30 m 3 d À1 oil-water separation continuous operation experiment was investigated for researching the reliabilities of the flotation column (Ø 400 mm × 4,000 mm) and the process. Water samples were taken every two hours, the water samples at the inlet and outlet were taken at the same time.
The circulating pressure was 0.16 MPa, the feeding rate was 1.25 m 3 h À1 , the powder coal consumption was 2 g coal g oil À1 , the frother dosage was 10 mg L À1 , the average SS concentration at the outlet was 25.56 mg L À1 .
The efficiencies of oil removal and SS removal were 99.14% and 76.88%, respectively. The two indexes meet the reinjection requirements of local oil deposits.

Results of process experiments
Coalescence-airflotation process on flotation column The circulating pressure was 0.16 MPa, the feeding rate was 1.25 m 3 h À1 , the frother dosage was 10 mg L À1 , experiment temperature was 20 W C. The experiment results are shown in Table 5.
As shown in Table 5, after the oily sewage was treated by the coalescence-airflotation process using the flotation column, the average oil concentration of the outlet samples was 263.05 mg L À1 , the average SS concentration of outlet samples was 47.33 mg L À1 . The efficiency of oil removal is more than 90% and SS removal efficiency reached 66.19%. So the coalescence-airflotation process by first using a flotation column in an industrial experiment was introduced and used to recycle the oil as much as possible.     The system of coalescence-airflotation-carrier preferential adsorption process on flotation column The circulating pressure was 0.16 MPa, the feeding rate was 1.25 m 3 h À1 , the powder coal consumption was 2 g coal g oil À1 , the frother dosage for second flotation column was 10 mg L À1 , the experiment temperature was 20 W C. The experiment results are shown in Figure 12.

Cost comparison
A cost comparison between the coalescence-airflotationcarrier preferential adsorption process and a conventional two-stage airflotation process is shown in Table 6. The following conditions were applied: treatment capacity of 30 m 3 d À1 , coagulant of poly-aluminium chloride consumption is 600 mg L À1 , flocculent of polyacrylamide consumption is 1 mg L À1 .
It can be seen from the Table 6, the cost of the coalescence-airflotation-carrier preferential adsorption process using a flotation column reduced the cost by 55.8% compared with a conventional two-stage airflotation process.
Results of 2,000 m 3 d À1 oil-water separation experiment The circulating pressure was 0.16 MPa, the feeding rate was 83.0 m 3 h À1 , the powder coal consumption of column 2 was 1 g coal g oil À1 , the experiment temperature was 20 W C. After the industrial experiment entered a stage of stable operation, the water samples were taken for four days. The experiment results are shown in Table 7.
As shown in Table 7, the average oil concentration of inlet samples was 1,017.18 mg L À1 , the average oil concentration of outlet samples through coalescence-airflotation process and carrier preferential adsorption process were 101.51 mg L À1 and 23.39 mg L À1 , respectively. The total oil removal efficiency was 97.70%. It can be seen from Figure 13, the bubble size of column 1 is smaller than column 2 because the powder coal which makes the small bubbles has a greater probability of forming bigger oilcoal-bubble complexes.
With the scaling up of the flotation column, all of the oil removal efficiencies were more than 95%, so the oil-water separation flotation column has good stability and oil removal efficiency. The flotation column using the  coalescence-airflotation-carrier preferential adsorption process provides a new and good way to treat the oily sewage.

CONCLUSIONS
The coalescence-airflotation-carrier preferential adsorption process and oil-water separation flotation column were introduced to solve the problem which can arise during polymer-oily sewage treatment, namely that at high oil concentrations, large numbers of fine oil droplets can result in serious emulsification. The 1, 30 and 2,000 m 3 d À1 oilwater separation systems were tested. The conclusions as follows.
The coalescence-airflotation-carrier preferential adsorption process is a novel water treatment process. This process arose from carrier flotation of mineral flotation.
Powder coal as a carrier adsorbs the oil droplets which include large, fine and emulsified oil droplets in oily sewage. Thereafter the oil-coal complexes can be separated from oily sewage by airflotation. The process contains two stages: the coalescence-carrier preferential adsorption and the airflotation separation stage.
The oil-water separation flotation column is novel equipment. It is integrated with cyclonic and flotation separation technology. In addition, the synergistic effect of multiple separation modes reinforces the separation effect and expands the range of oil concentrations in the oily sewage suitable for flotation separation.
According to the results of the 1, 30 and 2,000 m 3 d À1 oil-water separation system experiments, the coalescenceairflotation-carrier preferential adsorption process using a  flotation column has good reliability, stability and efficiency.
The oil removal efficiency is very good. In the 2,000 m 3 d À1 oil-water separation experiment, when the oil concentration of feed oily sewage is in the range 1,000 to 2,000 mg L À1 , the oil concentration and the oil removal efficiency at the outlet are 23.39 mg L À1 and 97.70%, respectively.
Finally, sediment is not produced during the oily sewage treatment using this novel process and flotation column. This is better in terms of both the environment and cost.
The cost was reduced by 55.8% compared with a conventional two-stage airflotation process.