-
E-mail
sievers.china@veolia.com
- Phone
-
Address
Building 5-6, Chuangqi Tiandi, 1761 Zhangdong Road, Pudong
Veolia Sievers analyzer
Successful correlation between BOD and TOC in ethylene wastewater process control
Date: 2025-09-22Read: 0
challenge
An ethylene production plant is seeking to improve the performance and monitoring of wastewater treatment processes. After the wastewater from the production equipment converges at the lifting station, it flows into the homogenization tank (EQ tank). Before the wastewater enters the Dissolved Air Flotation (DAF) system, operators add treatment chemicals to the water and adjust its pH value. The treated water is sent to the biological treatment system for further processing, and then clarified and discharged.
The factory needs to take samples at the discharge outlet every day and analyze them using Biochemical Oxygen Demand (BOD). BOD and other measurement data of water are used for compliance testingTry to calculate the total amount of organic matter emitted by the factory. However, the factory is unable to use the low BOD value reported as' Non Detect '. Another challenge is that BOD analysis requires a reporting time of 5 days, which makes it impractical to apply BOD analysis to monitoring and optimizing treatment processes.
Solution
The factory adoptsAnalyze monitoring planTo optimize water treatment processes to reduce organic emissions. Although the time requirement of BOD analysis makes this analysis method lose practical application value, the relationship between BOD and TOC can be used to establish the correlation between the two at each sampling point. Using these correlation coefficients for Total Organic Carbon (TOC) analysis, report almost real-time monitoring data, and predict the relevant biochemical oxygen demand (BOD) within minutesC，Biochemical Oxygen Demand-Correlated）” 数据。
The factory selects three sampling points to determine process operations and compares TOC and BOD data (see Table 1).
Sampling point location |
Process operation |
Upgrade Station |
Flow towards the homogenization tank and adjust the pH value |
Final clarification pool |
Reduce total suspended solids (TSS) |
Deep processing pool |
none |
Table 1: Wastewater Sampling Points
Due to the nonlinearity of BOD analysis data, it is required to separately derive the correlation coefficients of BOD and TOC samples at each sampling point.
Sampling multiple times a day can improve the accuracy of correlation. In this study, the factory monitored three testing points and extracted a total of seven samples within two weeks.
The first sampling point is located after the lifting station and before the homogenization tank, and the sample taken comes from a stable inflow. The measurement data is listed in Table 2.
Stable inflow | |||
TOC |
BOD |
BODC |
|
May 30th |
44.7 |
90.5 |
109.07 |
May 31st |
50.7 |
124 |
118.54 |
June 1 |
45.3 |
104.4 |
110.02 |
June 6 |
40.7 |
84.8 |
102.76 |
June 7 |
25 |
101 |
77.99 |
June 8 |
24.9 |
82.4 |
77.83 |
June 13 |
93.4 |
195 |
185.91 |
Table 2: Inlet data of homogenization tank
Any obvious outliers are replaced by the average value of BOD before and after, thereby increasing the correlation coefficient from 0.675 to 0.923. For wastewater, correlation coefficients higher than 0.5 can be used. The BODC values in Table 2 are calculated based on the relationship between measured BOD and TOC values.
BOD/TOC correlation
Figure 1: Correlation between BOD and TOC in the influent of the homogenization tank
The correlation between BOD and TOC in the inflow is very reliable, so TOC can be used as a substitute for BOD (see Figure 1).
The measurement results of the second sampling point at the outlet of the final clarification tank show that if the concentration is too low, the BOD value cannot be determined (see Table 3).
Water outlet | ||
TOC |
BOD |
|
May 30th |
5.97 |
3.79 |
May 31st |
6.85 |
3.39 |
June 1 |
8.19 |
3.95 |
June 6 |
1.08 |
Not detected |
June 7 |
7.33 |
3.83 |
June 8 |
7.7 |
Not detected |
June 12 |
6.76 |
4.23 |
Table 3: Final Clarifying Pool Data
Although the carbon content measured by TOC analysis has changed by 8 times, the sensitivity of BOD still does not meet the requirements for quantitative data. The BOD data in Table 3 shows that out of 7 samples, 2 samples could not be quantified and were reported as' undetected '. The data deviation between the other 5 BOD samples is within+/-4%, making it difficult to distinguish statistically. The BOD of the effluent can only be used for qualified/unqualified testing.
The BOD data of the deep processing pool (see Table 4) were reported as' undetected ', therefore it is not possible to establish a correlation with TOC. Although BOD is reported as' undetected ', TOC data is still accurate, precise, and linear.
Deep processing pool | ||
TOC |
BOD |
|
May 30th |
6.94 |
Not detected |
May 31st |
7.57 |
Not detected |
June 1 |
8.45 |
Not detected |
June 6 |
7.85 |
Not detected |
June 7 |
6.72 |
Not detected |
June 8 |
6.11 |
Not detected |
June 12 |
6.79 |
Not detected |
Table 4: Deep Processing Pool Data
Conclusion
This ethylene production plant has successfully utilizedTOC analysis methodTo monitor the wastewater treatment process. The inflow correlation coefficient they obtained is very reliable, so the almost real-time TOC analysis method can be used instead of the commonly used 5-day BOD testing method.
The organic carbon measurement results are the most reliable wastewater discharge data. The TOC analysis method can directly measure low ppm organic carbon in water, making it a more reliable monitoring and optimization tool. Operators can quickly respond to potential issues based on real-time data and take corrective measures in a timely manner.