Papermaking Wastewater: The Pollution Load of 90% Should Not Enter the Wastewater Treatment Plant

A pulp mill discharges three streams of water at the same time—black liquor, mid-stage water, and white water. The pollutants in black liquor account for more than 90% of the entire mill, but the place it should go is the alkali recovery boiler; white water has the largest volume and the lowest load, and the place it should return to is the paper machine; the only stream that truly needs to enter the biological tank is the mid-stage water. And the single figure of "water discharge per ton of pulp," in Table 1, Table 2, and Table 3 of GB 3544—2008, plus three footnotes, has a span of 10–80 t/t, and on top of that there are two different rulers pressing down: "oven-dry pulp" and "air-dry pulp (moisture 10%)." First make these two things clear, then discuss what process to use.

Manuscript Service · Industrial Water Treatment Technology Series · For industry technical personnel · All data are annotated with the original text of public standards, peer-reviewed literature, and public engineering sources

I. First distinguish "which stream it is": the three fates of the three streams

The pollution generation points of pulp and papermaking are distributed across three non-adjacent stages, and each has different water volume, concentration, and pollutant form. The table below lists the three streams side by side by stage. The first thing to look at is not COD comparison, but proportion—the pollution load of black liquor is an order of magnitude greater than that of the other two streams.

Read these three rows as three sentences:

First, the distribution of pollution load is completely out of proportion to the degree of water quality deterioration. White water is often the largest of the three streams by volume, but it carries the lowest pollution load; black liquor has the smallest volume (producing 1 t pulp generates roughly 10 t of black liquor), yet it accounts for more than ninety percent of the pollution. Sequencing projects on the principle of "treat the largest volume first" will spend investment in the wrong places.

Second, black liquor is the only wastewater for which "treatment equals recovery." The four stages of alkali recovery—black liquor extraction, evaporation, combustion, and causticization–lime recovery—remove the SS, COD, and BOD in black liquor all at once, while recovering alkali and generating secondary steam. The key lies in the phrase "all at once": it does not move pollutants into sludge or concentrate, but rather burns up the organics and converts the inorganics back into production raw materials. This is fundamentally different from all the end-of-pipe processes for the subsequent mid-stage water (coagulation, Fenton, ozonation, membranes)—the latter only transfer or transform, whereas the former destroys.

Third, the reason mid-stage water is "difficult" is not high concentration, but lignin. The COD of mid-stage water is often on the order of 800–2000 mg/L, much lower than that of pharmaceutical or coking wastewater; the problem is that a considerable portion of it is lignin and its degradation products, with B/C falling in the range of 0.20–0.35. This range sits right on the threshold line for Aerobic biological treatment of 0.3, so mid-stage water almost always requires a front-end Hydrolysis Acidification or Anaerobic stage to break the macromolecules apart first, rather than going directly into the Aerobic tank. The engineering ledger in Section 5 below will repeatedly show the same characteristic: only after the influent B/C is raised to 0.45–0.55 does the Aerobic stage begin to work properly.

II. The black liquor account is settled at the alkali recovery boiler: it is actually an alkali plant plus a thermal power plant

Alkali recovery is already highly mature on domestic wood pulp lines. The industry order of magnitude given by public engineering data is: black liquor extraction rate basically reaches 95%, with a maximum of 98%; alkali recovery rate exceeds 93%, with some reaching 98% or even 100%, and alkali is basically self-sufficient. The table below lists several alkali recovery systems reported publicly side by side, with attention focused on the last column—cost per ton of alkali.

What is truly worth remembering in this table is not the highest 99.3%, but the gap between the last two rows and the first four rows: the cost per ton of alkali for wood pulp lines is 517–654 元, while for straw pulp lines it is 973 元, self-calculated ratio 973 ÷ 517.3 ≈ 1.88. At the same time, the black liquor extraction rate for straw pulp is only 88%, and the alkali recovery rate is 79.1%—all three indicators lag behind simultaneously. This is not a matter of operational proficiency, but is determined by the raw material:

First, the black liquor from straw pulp has low solids calorific value and high silicon content. The silicon content of wheat straw and reed is much higher than that of wood. Once silicon enters the black liquor, it forms silicates that are difficult to melt in the recovery boiler, clogging the furnace and the smelt spout, forcing a reduction in firing concentration and load. Second, straw pulp lines are generally small in scale, and the thermal efficiency and automatic control level of the alkali recovery boiler decline with scale, while causticizing lime consumption is higher. Third, a low extraction rate means the black liquor that leaks out goes directly into the mid-stage wastewater — an extraction rate of 88% means that 12% of the black liquor enters the mid-stage wastewater, and the COD of black liquor is more than ten times that of the mid-stage wastewater, which is 12% enough to raise the influent COD of the mid-stage wastewater by an order of magnitude.

One statement in publicly available materials explains this chain very clearly: domestic wood pulp lines "basically all achieve a black liquor extraction rate of 95%, with the highest reaching 98%"; while straw pulp lines have "a black liquor extraction rate of 88%." The upper limit of the overall efficiency of alkali recovery lies not in the recovery boiler, but in extraction. This also explains why large domestic paper enterprises have continuously pursued "shifting the raw material structure from wheat straw to wood chips or logs" over the past two decades — the reports explicitly state the motivation for this reform as "addressing the problem that black liquor produced from wood chip pulp has a higher concentration than black liquor from wheat straw pulp, adopting internationally advanced technology to retrofit the black liquor extraction of the pulping system, achieving a black liquor extraction rate of over 95%."

There is also empirical evidence that in-process improvements are far cheaper than end-of-pipe treatment: Chenming carried out an oxygen delignification retrofit on the original bleaching process of its pulping line. After the retrofit, oxygen delignification was incorporated into black liquor combustion and no longer entered the mid-stage wastewater, reducing wastewater COD and SS concentrations by 15%, and reducing AOX by about 20%. In one sentence: Keeping pollutants in the black liquor to be burned is an order of magnitude cheaper than letting them enter the mid-stage wastewater and then paying to treat them.

III. Mid-stage Water: The Stream That Truly Requires Biological Treatment

The difficulty of mid-stage water lies not in concentration, but in two things: low biodegradability caused by lignin, and chlorinated organics brought by the bleaching stage. The order of magnitude from public sources is: COD load per ton of pulp around 310 kg, with B/C between 0.20 and 0.35. In engineering practice, this has therefore formed a fixed pre-treatment stage—first use physicochemical methods to remove coarse suspended solids and part of the lignin, then use Hydrolysis Acidification or Anaerobic processes to break down macromolecules, and finally Aerobic treatment plus advanced treatment.

The table below juxtaposes the engineering accounts of mid-stage water and combined wastewater obtained from public channels. Note the nature of the last two rows: the first two rows come from engineering companies' own statements (not peer-reviewed), and the middle row comes from an industry textbook paraphrase.

Two points in the four rows deserve separate mention:

First, the COD removal rate of the Hydrolysis Acidification stage is usually only around 10%, but it is the "switch" of the entire chain. In the industrial park case, the removal rate of the pretreatment stage is only 5%, and that of the Hydrolysis Acidification stage is only 10%, making it appear as though it is doing almost no work; yet it is precisely this stage that raises the B/C from 0.32 to 0.45, enabling the subsequent A/O process to achieve 77%. The bamboo pulp case is the same — Hydrolysis Acidification raises the B/C to 0.55, and only then can the oxidation ditch achieve a Total Nitrogen removal rate of 85%. Evaluating the Hydrolysis Acidification stage by "COD removal rate" will lead to the conclusion that "this tank is useless"; what it buys has never been removal rate, but biodegradability.

Second, color is the hardest hurdle in this chain. Public international data give the following order of magnitude: biological treatment can only remove 30%–50% of color, chemical coagulation adds another 60%–80% on top of that, and near-complete decolorization can only be achieved with Advanced Oxidation such as Ozonation or ozone/hydrogen peroxide. This corroborates the figures in the table above — the effluent color from the biological stage often remains in the hundreds (250 times after secondary settling in the industrial park case), the Fenton Oxidation stage brings color down from 250 to 30 in one pass, and the Ozonation stage further presses it down to 8.

IV. "Per ton of pulp" involves two different tons: oven-dry pulp and air-dry pulp differ by 11%

This is the most easily overlooked yet most conclusion-affecting definitional issue in this article. There are two sets of national standards in China that both govern water use in the papermaking industry, and their definitions of "one ton of pulp" are different:

— The footnote of GB 3544—2008 "Discharge Standard of Water Pollutants for Pulp and Paper Industry" states: pulp quantity is calculated as oven-dry pulp.
— The footnote of GB/T 18916.5—2022 "Norm of Water Intake — Part 5: Paper Products" states: the measurement unit of pulp is ton of air-dry pulp (moisture content 10%).

In other words, the same "50 t/t" refers to per ton of oven-dry pulp in the discharge standard, and per ton of air-dry pulp in the water intake norm. 1 air-dry ton of pulp = 0.9 oven-dry ton of pulp, a difference of 11.1%. This difference may look small, but as long as the two figures are put together for subtraction, the conclusion will be noticeably skewed.

Here is a comparison that can be calculated through in one pass: for a certain newly built pulp mill, the benchmark discharge volume per unit product is taken as the pulping enterprise tier in Table 2 of GB 3544—2008, 50 t/t (oven-dry pulp basis), and the water intake norm per unit product is taken as the bleached chemical wood pulp tier in Table 2 of GB/T 18916.5—2022, 60 m³/t (air-dry pulp basis).

Self-calculation process: Convert the discharge volume to the air-dry pulp basis — each 1 air-dry ton of pulp contains 0.9 oven-dry ton of pulp, so the discharge volume is 50 × 0.9 = 45 m³/air-dry ton; water intake is 60 m³/air-dry ton; the difference between the two is 60 − 45 = 15 m³/air-dry ton. If no conversion is made and subtraction is done directly, the result is 60 − 50 = 10.15 versus 10, error 50%.This 15 m³ is the water lost to evaporation and contained in products and sludge — counting it as 10 would directly underestimate the "water of unknown destination" in the water balance by more than one-third, thereby affecting the estimation of Reclaimed Water Reuse scale and concentrated water volume.

There is a second layer to note here: the water intake quota itself is also changing.The 2022 version divides the indicators into three tiers: existing, new (expansion/renovation), and advanced, with considerable differences for the same product; while the 2012 version has only two tiers: existing and new. Taking bleached chemical wood pulp as an example: the 2012 version has existing 90, new 70; the 2022 version has existing 75, new 60, advanced 50.Over ten years, the "existing enterprise" tier has tightened from 90 to 75, with the advanced value at only 50, a 1.5-fold difference among the three.The table below is the complete comparison of the three tiers in the 2022 version, used to illustrate the tiered benchmarks within the same standard.

The comparison between the last row and the second-to-last row is worth noting: the most lenient water intake quota is for "Bleached Chemical Non-wood Pulp" at 110 m³/t, while the strictest is for "Corrugating Medium" at 12 m³/t (advanced value), a difference of 9 times within the same standard. This span is not a contradiction within the standard, but rather the product process determines the water volume: non-wood pulp requires silicon removal, multiple washing stages, and black liquor extraction is difficult; corrugating medium uses waste paper pulp, and the water in the papermaking process can be highly closed-loop recycled. Using an industry average such as "water intake per unit product in the paper industry" to evaluate any specific factory is invalid.

V. "Baseline Discharge Volume per Unit Product" Is a Lookup Table, Not a Single Number

The compliance determination of GB 3544—2008 consists of two parts: concentration limits (three tables) + baseline discharge volume per unit product (also a tiered system). This section breaks down the tiers. The applicable relationship of the three tables is: Table 1 is for transitional use by existing enterprises that went into operation before 2008 年 (2009-05-01 to 2011-06-30), Table 2 is implemented by new enterprises from 2008-08-01 and by existing enterprises from 2011-07-01, and Table 3 is the special emission limit, applicable to designated areas with high land development density and small water environment capacity. The table below lists only two items, COD and baseline discharge volume, and places the four categories of enterprises side by side.

In the three tables, the baseline effluent volume for pulp mills is 80 → 50 → 30, and for paper mills is 20 → 20 → 10. However, the footnotes of GB 3544—2008 add another layer of "tier-shifting" rules on top of these three tiers, and the directions are not even consistent:

—Footnote to Table 1: When the output of bleached non-wood pulp accounts for more than 60% of the enterprise's total pulp consumption, the benchmark water discharge per unit product is 80 t/t (the same as the benchmark value for pulping enterprises).
—Footnote to Table 2: When bleached non-wood pulp accounts for more than 60%, the benchmark water discharge is 60 t/t; when self-produced waste paper pulp accounts for more than 80% of the enterprise's total pulp consumption, the benchmark water discharge is 20 t/t.
—Footnote to Table 3: When self-produced waste paper pulp accounts for more than 80%, the benchmark water discharge is 15 t/t.

Look at this row in Table 2: the benchmark value for pulping enterprises is 50; the same 50 is pulled in two directions by two footnotes—the one where bleached non-wood pulp accounts for more than 60% (straw pulp line) is relaxed to 60 (relaxed by 20%), and the one where self-produced waste paper pulp accounts for more than 80% (waste paper line) is tightened to 20 (tightened by 60%).This is not a contradiction; it acknowledges the objective water volume difference among different raw materials: straw pulp requires silicon removal and more washing stages, while waste paper pulp basically does not generate black liquor.But it brings an operational reminder: before reporting water discharge, you must first determine which tier you are in, and "which tier" depends on the pulp type structure in the environmental impact assessment document or production statistics, not on what the enterprise claims.The footnote also states: when verifying the actual water discharge per unit product of an integrated pulping and papermaking enterprise, the basis is the sum of the enterprise's pulp output and the amount of purchased market pulp—purchased pulp must also be included in the denominator.

The standardized determination uses the same one-sided formula:

C基 = Q总 ÷（Σ Yi × Qi基）× C实

Where Q总 is the total water discharge, Yi is the output of the i-th product, Qi基 is the benchmark water discharge per unit product for that product, and C实 is the measured concentration. The additional rule is: if the ratio of Q总 to Σ（Yi·Qi基）is less than 1, the measured concentration shall be used as the basis for determination.In other words, when actual water discharge is below the benchmark, there is no conversion or reduction; when it exceeds the benchmark, the measured concentration is amplified by a multiple—a mechanism that only penalizes and does not reward (the parallel TOC, acute toxicity, and other parameters have already been similarly dissected in the pharmaceutical wastewater section of 10-02).

Finally, there is the "supplementary role" of local standards. GB 3544—2008 and the six pharmaceutical standards share one common formulation: no indirect discharge limits are set at the national level.The original text of the standard only states that "when a construction project intends to discharge wastewater into a municipal drainage system with a sewage treatment plant, the construction unit and the municipal sewage treatment plant shall implement it in accordance with the provisions of the preceding paragraph." The DB35/1310—2013 "Water Pollutant Discharge Standard for the Pulping and Papermaking Industry" issued by Fujian Province in 2013 年 (existing enterprises shall comply starting from 2014-01-01) handles this matter in two different ways: first, it splits the benchmark water discharge from a single line into a three-dimensional lookup table; second, it explicitly writes indirect discharge into the clauses.

Compare the range with the national standard: Pulping enterprises go from a single value of 50 to a range of 10–50, papermaking enterprises from 20 to 5–20, and integrated pulp and paper enterprises from 40 to 12–35; the water pollutant discharge concentration limits basically follow the national standard, with only the COD for waste paper papermaking tightened from 90 mg/L to 80 mg/L. Meanwhile, DB35/1310—2013 explicitly specifies indirect discharge requirements: Before discharge into urban sewage treatment plants, the concentration and discharge volume shall comply with Table 1 and Table 2; for indirect discharge into other public sewage treatment systems, GB 3544 and relevant national standards shall apply. This sentence fills the gap of "to be determined by the enterprise and the sewage treatment plant" mentioned earlier—Whether a province's discharge criteria are stringent does not depend on how tough its concentration limits are, but on whether it has broken down the baseline discharge volume in detail and locked down indirect discharge.

VI. AOX and Dioxins: The Only Two Indicators at the "Workshop Outlet"

GB 3544—2008 Table 2 and Table 3 contain a total of 10 pollutants, of which only two — Adsorbable Organic Halogens (AOX) and Dioxins — have their monitoring location set at the "workshop or production facility wastewater discharge outlet", while the remaining 8 items (pH, color, Suspended Solids, BOD₅, COD, ammonia nitrogen, Total Nitrogen, Total Phosphorus) are all at the "enterprise wastewater main discharge outlet". The AOX limit values are 12 mg/L in Table 2 and 8 mg/L in Table 3; the dioxin limit values are 30 pgTEQ/L in both tables, and the standard stipulates that the dioxin indicator shall be monitored once per year. The footnote also states: AOX and dioxins apply only to "cases where chlorine-containing bleaching processes are used".

These three things together form a pattern that is easily misread: the limit values are tied to the workshop outlet, yet the measured values are often taken from the main discharge outlet — the two numbers are simply not the same thing. Publicly available peer-reviewed measured data make this discrepancy very clear.

Looking at the two rows together, three conclusions can be drawn:

First, AOX cannot be solved in one pass by biological treatment; the high-molecular-weight half essentially remains. Data from a North American 8 plant show that biological treatment can remove more than half of the low-molecular-weight AOX, but the high-molecular-weight fraction remains between 1% and 47% and is independent of process type. This means that compliance for AOX does not depend on which process the wastewater treatment station chooses. The accounts of Company A illustrate this best: the Aerobic stage contributes only 35.9%, 59.8% of the AOX is achieved by the downstream Advanced Oxidation stage—by its own calculation 35.9% + 59.8% = 95.7%, meaning that more than half of this plant's AOX removal is obtained by dosing Fenton reagents per ton of water. By contrast, because Company B's Anaerobic stage has already degraded 90% of the organic matter, the combined Anaerobic plus Aerobic stages achieve 94.4%, and the Advanced Oxidation stage only needs to treat soluble COD 19 kg/d to meet the standard. For the same parameter and the same limit, the reagent consumption in the Advanced Oxidation stage of the two plants differs by an order of magnitude, and the difference comes from how well the preceding Anaerobic stage performs.

Second, the real leverage for AOX is in the bleaching workshop. Public information gives the following magnitudes: elemental chlorine bleaching (conventional process) generates about 0.5–3.0 kg of AOX per ton of pulp; after switching to the chlorine dioxide-based ECF (Elemental Chlorine Free) process, typical values are 0.5–1.5 kg/t, with the EU BAT-AEL level at 0.25 kg/t; dioxin and furan emissions decrease by more than 99% relative to the conventional elemental chlorine process, with most falling below the analytical detection limit. The completely chlorine-free TCF process produces no AOX at all, at the cost of increased oxygen, hydrogen peroxide, and ozone consumption and greater carbohydrate degradation, with effluent COD actually higher than that of ECF. A corresponding domestic case is the previously mentioned Chenming oxygen delignification retrofit: after oxygen delignification was incorporated into black liquor combustion and no longer entered the mid-stage wastewater, wastewater COD and SS decreased by 15% and AOX decreased by about 20%.

Third, the footnote determines applicability; plants without chlorine bleaching should not report these two parameters. The standard footnote explicitly limits AOX and dioxins to "cases where chlorine bleaching processes are used." For a plant that fully uses ECF or even TCF, using "our AOX is not detected" as an achievement does not hold up—because this parameter simply does not apply to its process stage; conversely, for a plant still using elemental chlorine bleaching, taking the AOX figure at the final outfall to benchmark against the 12 mg/L limit at the workshop outlet is also comparing two different bases.

VII. White Water: The Cheapest and Most Easily Overlooked of the Three Water Streams

The treatment target of whitewater is Suspended Solids / SS, not COD; its output is reclaimed water plus recovered fiber, not compliant discharge. This logic determines that its cost-effectiveness far exceeds the other two streams of water. The industry level given by public engineering data is: the whitewater reuse rate in the general papermaking industry is around 83%; after converting air flotation to multi-disc filtration, a household paper enterprise reached 91%, achieved 100% recovery of pulp, and reduced wastewater discharge by 10 万 t and solid waste by 4000 t per year.

The table below juxtaposes publicly available whitewater recovery and plant-wide reuse data. The last three rows come from the list of typical cases of industrial wastewater recycling released by the industrial and information technology department (2024 年), which belongs to government platform data.

The most worthwhile method to borrow from this table is the last row: the white water account must be calculated on two counts together — how much is saved per ton of water + how much pulp is recovered per ton.Calculate it yourself — recovered water is cheaper than fresh water by 0.11 元/m³, which is not a large figure in itself; but the daily recovery of oven-dry pulp 6 t, at 2000 元/t, amounts to 1.2 万 yuan/d, which annualized at 330 d/a is approximately 396 万 yuan/a, two orders of magnitude higher than the water fee difference.The economics of a white water recovery project come mainly from fiber, not from water.This is exactly the opposite direction from the conclusion in the 09-30 oily wastewater article that "dissolved air flotation only moves oil from water into sludge, and the float sludge disposal cost is more than ten times the electricity consumption": the "float sludge" of white water is pulp that can be sold for money.

There is one more scope issue that needs to be raised separately: for the term "water consumption per ton of paper," which type of paper is used as the denominator makes a huge difference. In the data disclosed by the same company in the same year, clean water consumption per ton of machine-made paper (including pulp) was 13.2 m³, but when broken down, coated paper was 2.9, light-coated paper was 3.1, white card paper was 4.4, and chemical mechanical pulp was 3.6 m³ per ton of pulp—the highest and lowest differ by 4.5 times (13.2 is the comprehensive scope including pulp, while under the single-product scope, coated paper is the lowest at 2.9). Combining this with the two scopes in Section 4, this article contains a total of three denominators all called "ton": ton of oven-dry pulp (baseline discharge volume for pollutant discharge standards), ton of air-dry pulp (water intake quota), and ton of paper (corporate water consumption reports). The three "tons" are not equal to one another, and before comparing any two figures, the denominator must first be confirmed.

VIII. Comparison of Three Pathways and Five Engineering Truths

Let us condense the entire article into one selection table. The three streams of water are not "different degrees of treatment," but rather three completely different professional fields: black liquor is chemical recovery, mid-stage water is biochemical + Advanced Oxidation / AOP, and white water is solid-liquid separation and reuse.

Five Engineering Truths

1. The ceiling of alkali recovery lies in extraction, not in the recovery boiler. For wood pulp lines, the extraction rate is 95%, with a maximum of 98%, and the recovery boiler can achieve a recovery rate of 93%–99.3%; for straw pulp lines, the extraction rate is only 88%, so the recovery rate can only reach 79.1%.For every one percentage point drop in extraction rate, a batch of high-concentration black liquor goes directly into the mid-stage wastewater—black liquor COD is more than ten times that of mid-stage wastewater, and the treatment cost for this portion must be paid at the mid-stage wastewater process rate, far exceeding the investment at the extraction stage.

2. There is no "small-scale economy" in alkali recovery. The gap between the cost per ton of alkali for wood pulp at 517–654 元 and straw pulp at 973 元 is 1.88 times, and straw pulp must additionally endure furnace condition problems caused by silicon interference. For small and medium-scale non-wood pulp lines, rather than forcing in alkali recovery, it is better to first evaluate raw material substitution (the approach publicly reported by multiple enterprises is "changing the raw material structure entirely from wheat straw to wood chips or logs" while simultaneously carrying out black liquor extraction retrofits).

3. The difficulty of mid-stage wastewater is not in COD but in color, and the only way out for color is through staged combinations. Biological treatment can only remove 30%–50% of color, coagulation adds another 60%–80%, and near-complete decolorization depends on ozone or ozone/hydrogen peroxide. The correct engineering approach is to let each stage do what it is good at: coagulation first removes suspended lignin, hydrolysis acidification/anaerobic raises B/C, aerobic consumes the biodegradable portion, and Fenton or ozone finally treats dissolved refractory color—do not expect a single stage to bring down color in one pass.

4. The applicability of AOX and dioxins must be judged by "whether chlorine bleaching is involved," and the total outlet figures cannot be used to benchmark against workshop outlet limits. The standard footnotes explicitly define the applicable conditions, and the limit monitoring point is at the workshop outlet. At the same time, two-year actual measurements at 8 plants in North America show that biological treatment removes only 34%–46% of AOX, with the high molecular weight fraction hardly decreasing—expecting the wastewater treatment plant to bring AOX into compliance is the wrong direction.

5. The three "tons" cannot be directly subtracted from one another. Ton of oven-dry pulp (GB 3544 baseline discharge volume), ton of air-dry pulp containing 10% water (GB/T 18916.5 water intake quota), and ton of paper (enterprise water consumption report) are three different denominators, and within the same water intake quota there are still three tiers—existing/new/advanced—and within the same enterprise different paper grades can differ by 4.5 times.Whenever a water balance check "does not add up," first go back and check the denominator, then suspect the meter.

Two Work Sequences That Can Be Directly Implemented

Troubleshooting sequence for new or retrofit projects: ① First check the pulp type structure and bleaching process (determines which tier of baseline discharge volume to consult, and whether AOX and dioxins apply); ② Then verify whether the black liquor extraction rate reaches the 95% order of magnitude (determines the influent load of mid-stage wastewater); ③ Then use the ratio ρ基 of actual discharge volume to baseline discharge volume to make a compliance pre-judgment (when the ratio is greater than 1, the measured concentration will be amplified); ④ Only then arrange the process chain for mid-stage wastewater, and it must include hydrolysis acidification or anaerobic treatment to raise B/C above 0.45.

Cost-saving sequence for operating projects: white water recovery retrofit (lowest investment, recovered fiber directly monetized) → black liquor extraction stage optimization (every 1 percentage point increase reduces downstream treatment load) → bleaching process retrofit (oxygen delignification, ECF, directly eliminating AOX and COD at the source) → chemical optimization of the advanced treatment stage for mid-stage wastewater (act on this last, because after the previous steps are done well, chemical consumption in the advanced stage will automatically drop by an order of magnitude—the difference between Enterprise A and Enterprise B is exactly this order of magnitude).

References

1. GB 3544—2008 "Water Pollutant Discharge Standard for the Pulp and Paper Industry", issued by the Ministry of Environmental Protection and the General Administration of Quality Supervision, Inspection and Quarantine on 2008-06-25, implemented on 2008-08-01; simultaneously repealing GB 3544—2001 and Huan Fa [2003] No. 152. This standard was first published in 1983 年, first revised in 1992 年, and second revised in 2001 年.
2. GB 3544—2008 Table 1 (Water Pollutant Discharge Limits for Existing Enterprises), Table 2 (Water Pollutant Discharge Limits for New Enterprises), Table 3 (Special Water Pollutant Discharge Limits), and the notes under the three tables (AOX and dioxins apply to processes using chlorine-containing bleaching; pulp quantity is calculated as oven-dry pulp; the actual water discharge per unit product for integrated pulp and paper enterprises is verified based on the sum of pulp production and purchased market pulp; baseline water discharge adjustments for bleached non-wood pulp proportion greater than 60% and self-produced waste paper pulp proportion greater than 80%).
3. GB 3544—2008 conversion formula for water pollutant discharge concentration based on baseline water volume (C_base = Q_total ÷ Σ(Y_i·Q_{i base}) × C_actual) and the supplementary rule that "when the ratio is less than 1, the measured concentration shall be used as the basis for determination".
4. GB/T 18916.5—2022 "Water Intake Quota Part 5: Paper Products", Table 1 (Water Intake Quota for Existing Paper Enterprises), Table 2 (Water Intake Quota for New and Expanded/Reconstructed Paper Enterprises), Table 3 (Water Intake Quota for Advanced Paper Enterprises), and footnotes such as "the measurement unit for pulp is tons of air-dry pulp (moisture content 10%)".
5. GB/T 18916.5—2012 "Water Intake Quota Part 5: Paper Products" (implemented on 2013-01-01), water intake quotas for existing/new enterprises, used for version comparison with the 2022 edition.
6. DB35/1310—2013 "Water Pollutant Discharge Standard for the Pulp and Paper Industry", issued by the Fujian Provincial Bureau of Quality and Technical Supervision and the Fujian Provincial Department of Environmental Protection; Table 1 (Direct Water Pollutant Discharge Limits), Table 2 (Baseline Water Discharge per Unit Product), and Articles 4.3, 4.4 regarding direct discharge and indirect discharge provisions.
7. "Fujian Introduces Stricter-Than-National-Standard Discharge Standards for the Paper and Cement Industries", China Economic Net Local Economy Channel, 2013-08-05 (pulp enterprises 50 → 10～50, paper enterprises 20 → 5～20, integrated pulp and paper enterprises 40 → 12～35; waste paper papermaking COD 90 → 80 mg/L).
8. "Fujian Provincial Bureau of Quality and Technical Supervision and Fujian Provincial Department of Environmental Protection Issue Water Pollutant Discharge Standard for the Pulp and Paper Industry (Abstract)", Journal of Fujian Paper Industry Association (Tables 1 and 2 reproduced in full).
9. "AOX Migration and Occurrence in Pulp and Paper Wastewater Treatment Systems", China Pulp & Paper, Issue 6 of 2023 年 (Enterprise A eucalyptus bleached pulp AOX removal 22.6 kg/d, aerobic stage 35.9%/advanced stage 59.8%; Enterprise B bagasse AOX removal 109.5 kg/d, anaerobic plus aerobic 94.4%; AOX at both enterprises' final outlets reduced to below 1.0 mg/L).
10. Bryant, C. W., Avenell, J. J., Barkley, W. A. & Thut, R. N. The Removal of Chlorinated Organics from Conventional Pulp and Paper Wastewater Treatment Systems. Water Science & Technology, 1992, 26(1-2): 417 (two-year AOX removal tracking of 8 full-scale systems in North America: activated sludge average 46%, aerated stabilization basin average 34%, best 53%).
11. "Treatment and Resource Recovery of Pulp and Paper Wastewater", Wastewater Treatment Engineering Network (dowater.com) technical literature (black liquor pollutants account for over 90% of total pollution discharge from the paper industry; middle-stage water accounts for 8%～9%, COD load per ton of pulp approximately 310 kg, B/C 0.20～0.35; white water composition is mainly insoluble COD; the four-stage alkali recovery can simultaneously remove SS, COD, BOD and recover alkali and secondary steam).
12. "Paper Black Liquor Treatment Methods", Wastewater Treatment Engineering Network, 2018-02-24 (for every 1 t of kraft pulp produced, approximately 1 t of organic matter and 400 kg of alkali and sulfides dissolve in black liquor; comparison of black liquor treatment technology routes).
13. "Paper Wastewater Treatment Engineering Solutions", Fujian Paper Industry Association (middle-stage water COD 800～2000 mg/L, BOD 200～500 mg/L, lignin 50～200 mg/L, pH 7～8; producing 1 t of pulp generates approximately 10 t of black liquor; black liquor impurities account for 10%～20%, approximately 1/3 inorganic sodium salts, 2/3 organic alkali lignin and hemicellulose fatty acid resin sodium salts; white water BOD is typically only tens of mg/L).
14. "Special Boilers" industry knowledge base compilation (around 2013 年), section on current status of alkali recovery technology development: black liquor extraction rate basically reaches 95%, maximum 98%; alkali recovery rate exceeds 93%, individual cases 98% to 100%; recovery rates and cost per ton of alkali for three alkali recovery boilers 1100 t/d, 5500 t/d, 6500 t/d (654 元, 517.30 元); straw pulp line extraction rate 88%, alkali recovery rate 79.1%, cost per ton of alkali 973 元.
15. "93 People, Nearly 4000 万 Yuan: This Workshop Turns Paper Black Liquor into Treasure", Xinhua News Agency, 2026-05-18 (Yunnan Yunjing Forestry & Pulp Mill alkali recovery rate above 95%, annual savings of 8000 t standard coal, cumulative cost reduction and efficiency improvement of nearly 4000 万 yuan).
16. "Yulong Paper Fully Treats Wastewater", Shandong Provincial Department of Ecology and Environment, 2006-11-16 (75 t/d wood pulp alkali recovery project alkali recovery rate 85%; black liquor extraction rate above 95%; water consumption per ton of pulp decreased by 50%; white water closed-loop treatment and recycling).
17. "Chenming Paper 2010 年 Social Responsibility Report", China.com Finance (black liquor extraction rate above 95%; after oxygen delignification incorporated into black liquor combustion, wastewater COD and SS reduced by 15%, AOX reduced by approximately 20%; alkali recovery rate above 90%; daily treatment of 450 t self-produced pulp black liquor with 3000 kWh power generation units; daily treatment of 13.5 万 m³ middle-stage water; papermaking white water reuse rate above 90%; clean water consumption per ton of machine paper including pulp 13.2 m³, coated paper 2.9, light-coated paper 3.1, white card paper 4.4, chemi-mechanical pulp 3.6 m³).
18. Pulp & Paper Water Treatment industry technical review (Milton Roy public technical page): comprehensive wastewater COD 1500～5000 mg/L, B/C 0.3～0.5, color removal magnitude (biological treatment 30%～50%, coagulation plus additional 60%～80%, ozone near-complete decolorization), elemental chlorine bleaching AOX 0.5～3.0 kg/t pulp / ECF 0.5～1.5 kg/t / EU BAT-AEL 0.25 kg/t.
19. Bleaching of wood pulp (pulp bleaching process review entry, including ECF reducing dioxin and furan emissions by over 99% relative to elemental chlorine processes, with most reduced to below detection limits; ECF accounts for 99% of bleached chemical pulp production in North America; AOX emissions reduced by over 90%; modern pulp mills achieve over 80% water reuse through internal recycling of white water and filtrate; comparative study showing TCF process COD higher than ECF).
20. "2024 年 Typical Cases of Industrial Wastewater Recycling" (Vinda Paper (Zhejiang) Co., Ltd.: white water reuse rate above 95%, fiber reuse rate 100%, water consumption per unit product 7.56 m³/t, water reuse rate 99.54%; Asia Symbol (Shandong) Pulp & Paper Co., Ltd.: water reuse rate 95.93%, water consumption per unit product 12.52 m³/t, reclaimed water utilization 666.28 万 m³/a; Henan Jianghe Paper Co., Ltd.: water reuse rate above 90%).
21. "Developing Circular Economy, Taking the Path of Sustainable Development", Fujian Paper Industry Association (white water graded recovery retrofit: newsprint 15 → 10, cultural paper 17 → 14, waste paper deinking pulp 30 → 7, wood chip groundwood pulp 30 → 18, chemical pulp 110 → 61 m³/t; paper machine white water 100% closed-loop circulation; alkali self-sufficiency rate 100%, alkali recovery rate above 93%).
22. "Vinda: 'Reclaimed Water Reuse' System Treats Wastewater, Saving Water and Being Environmentally Friendly", China Energy Conservation Network (white water recovery changed from air flotation to multi-disc filtration: white water reuse rate 83% → 91%, pulp recovery 100%, annual pollutant reduction 10 万 t, solid waste 4000 t; one plant produces approximately 8 万 t of household paper annually, annual water savings after reclaimed water reuse 10 万 t).
23. "CQF Type High-Efficiency Ionized Air Flotation: The Best Choice for Water Conservation in the Paper Industry", equipment supplier public technical materials (non-peer-reviewed; Guangzhou Paper No. 4 Paper Machine white water recovery: influent SS ≤3234 mg/L → effluent ≤45 mg/L, removal rate ≥98.5%; COD ≤639 mg/L, removal rate ≥81.8%; water consumption per ton of paper reduced to below 30 m³/t; white water recovery cost 0.259 元/m³ vs. clean water 0.369 元/m³, daily recovery of oven-dry pulp 6 t at 2000 元/t yields daily savings of 1.2 万 yuan).
24. "6000 Characters! Let You Understand 'Hydrolysis Acidification Process'", China Water Industry Network (reproduced from the 15th Sludge Conference materials), paper wastewater hydrolysis acidification-aerobic case (influent COD 4120 mg/L, BOD₅ 1630 mg/L, SS 2080 mg/L, hydrolysis reactor HRT 5.8 h, effluent COD 354, BOD₅ 92.5, SS 95.6 mg/L, removal rates 91.4%/94.3%/95.4%); this manuscript has fact-checked the inconsistency between this case's B/C = 0.396 and its "poor biodegradability" statement. Also cites hydrolysis acidification data from printing and dyeing, pharmaceutical, and brewery wastewater for cross-comparison.
25. "High COD Paper Wastewater Treatment Case", "Bamboo Pulp Paper Mill Wastewater Treatment Case", "Paper Industrial Park Wastewater Treatment Case", Weilin Environmental / Xinlin Environmental public engineering cases (engineering company self-reported, non-peer-reviewed; the three engineering data sets and cost breakdowns cited in Sections 3 and 8 of the main text are all from these three cases).
26. Smurfit Westrock Argentina Coronel Suárez Paper Mill 2025 年 water treatment system upgrade public statement (internal circulation anaerobic stage placed before the original aerobic system, reducing COD entering the aerobic stage by 75%; 2024 to 2025 年 COD reduction 73%, equivalent to approximately 106 t COD).
27. "Luzhou Municipal Ecology and Environment Bureau Pollution Source Supervisory Monitoring Results Disclosure" (actual implementation of GB 3544—2008 Table 2 in pollutant discharge permits and supervisory monitoring: integrated pulp and paper enterprises baseline water discharge per unit product 16.1 m³/t, COD 30 mg/L; pulp enterprises baseline water discharge 38.4 m³/t, COD 48 mg/L, measured values all below limits).
28. "Water Pollutant Discharge Standard for the Pulp and Paper Industry GB 3544—2008" standard text appendix (list of pollutant analysis method standards, including GB/T 15959—1995, HJ/T 83—2001 adsorbable organic halogens determination, HJ/T 77—2001 dioxin determination).

Manuscript Express · Industrial Water Treatment Technology Series | All data in this article are from the above public standard texts, peer-reviewed literature, and public engineering sources; equipment supplier and environmental company self-reported data have been individually marked as "non-peer-reviewed" in the main text and tables, for directional reference only, and should be verified through bench-scale or pilot-scale testing before engineering application. All values calculated by this manuscript have been marked as "self-calculated" with formulas provided. The current validity of GB 3544—2008, the implementation status of DB35/1310—2013, and the COD load per ton of pulp 310 kg/t marked as [To Be Verified] in this manuscript are recommended to be verified against the standard texts and local ecology and environment authorities before formal citation.