你以为的不一定是真的——《对膜通量减少的研究发现了一个不太可能的原因》

DOI：10.1002/opfl.1308

原文作者：Erin K Moore

原文出处：J Opflow

翻译：阮辰旼

Decreased Filter Run Volume Investigation Leads to an Unlikely Culprit

对过滤通量减少的研究发现了一个不太可能的原因

The filters at Poughkeepsies’ Water Treatment Facility (PWTF) in New York experienced lower unit filter run volumes (FRVs) since early 2017, when they became biologically active. The plant has a design capacity of 19.3 mgd, but the biofilters limited its capacity to about 12 mgd. To operate at capacity, the plant investigated its filtration processes to determine the cause of its low FRVs.

PWTF treats 10.4 mgd of Hudson River water for the town and city of Poughkeepsie. The city constructed the first water treatment facility in the United States in 1872 using slow-sand filters. With the addition of chlorine in 1909, the plant ran for 90 years before a new plant replaced it in 1962.

At that time, the PWTF treatment process included chemical addition, solids contact clarifiers, sedimentation, and conventional filtration. Ultraviolet (UV) disinfection was installed, and the filters—including underdrains and air backwash—were upgraded in 2004. Although the plant was operating well, a new water quality issue emerged caused by disinfection byproducts (DBPs), which were produced by conditions in the distribution system, not by the raw water source (the Hudson River).

纽约Poughkeepsies水厂（PWTF）自2017年初使用生物过滤好，开始出现单位过滤通量（FRVs）降低的情况。该水厂的设计能力为19.3百万加仑每天，但由于生物滤池的原因，其实际生产能力被限制在仅约12百万加仑每天。为了让水厂恢复到设计生产能力运营，该水厂仔细调查研究了整个过滤工艺过程，以确定其单位过滤通量（FRVs）变低的原因。

Poughkeepsies水厂（PWTF）为波基普西镇的城镇地区处理10.4 百万加仑每天的哈德逊河原水。该市曾在1872年建造了美国第一个水处理设施，使用慢滤池工艺，随后1909年引入了氯气消毒工艺。该厂一共运行了90年，直到1962年一个新的水厂取代了它。

当时，Poughkeepsies水厂（PWTF）的处理工艺包括化学药剂投加、接触澄清、沉淀和传统的过滤。2004年该水厂安装了紫外线（UV）消毒，并对过滤工艺，包括排水和空气反冲洗进行了升级。虽然水厂的运行情况良好，但出现了一个由消毒副产物（DBPs）引发的新的水质问题，这些消毒副产物是因输配系统的环境条件产生的，而不是来自哈德逊河原水。

DBPs form when the chlorine residual that controls pathogens in the water distribution system has been in contact with trace natural organics, forming compounds that are hazardous to human health. The US Environmental Protection Agency (USEPA) began regulating these compounds with the Stage 1 and 2 Disinfectants and Disinfection Byproducts Rules (DBPRs); facilities where DBPs were detected were given compliance dates.

输配系统中，当用于控制病原体的余氯与微量的天然有机物接触时，就会形成消毒副产物，一种对人类健康有害的化合物。美国环境保护总署（USEPA）开始制定和执行消毒剂和消毒副产品规则（DBPRs Stage 1 and 2）来管控这些化合物，被检测到存在消毒副产物的水厂或供水设施都有参照DBPRs规则执行的最后期限。

To comply with the DBP regulations, Poughkeepsies’ Joint Water Project Board installed an ozone treatment system. The advantage of using ozone is its ability to break down the natural organics that can result in DBPs forming in the distribution system. However, the remaining degraded organics leaving the ozone system need enhanced filtration or they can increase biology in the distribution system. Biologically active filters that feed on organic nutrients located after the ozone system remove these organics before UV disinfection and chlorine addition.

为了遵守DBPRs规则的要求，Poughkeepsies水厂的联合水项目委员会安装了一个臭氧深度处理系统。使用臭氧的好处是它能够分解可能导致消毒副产物在输配系统中生成的天然有机物。然而，臭氧氧化后的剩余的被降解有机物还需要通过强化过滤去除，否则它们会增加输配系统中的生物量。在臭氧氧化之后的工艺是以有机营养物为食的生物活性炭滤池，生物滤池能够帮助在紫外线消毒和加氯消毒工艺之前实现去除这些有机物。

In 2015, ozone contactors were installed prior to the filters. Also, at that time, the sand and anthracite filter media were removed and replaced with granular activated carbon (GAC) filters. The filters started to become biologically active in September 2016 when the upstream ozone process was placed online. In January 2017, operating adjustments were made to optimize organic removal.

Initially, lowered FRVs were attributed to hot summer temperatures, new operation and maintenance procedures for plant staff, and anticipated reductions known to be associated with biofiltration use. However, when FRVs in the summer of 2017 continued to decrease (Figure 1), the issue became a significant concern.

Under normal summer conditions before the conversion to biofiltration, the plant would see FRVs of at least 30,000 gal/ft2; however, FRVs began to significantly decrease to below 10,000 gal/ft2 during a time when the plant should be doing its best. While some FRV reduction was anticipated, this level of reduction was unexpected and made it impossible for the plant to reach its design-rated capacity of 19.3 mgd.

2015年，该水厂首先在滤池之前安装了臭氧接触器。与此同时，原来滤池中的沙子和无烟煤过滤介质被移除，并被颗粒活性碳（GAC）取代。2016年9月，当前端的臭氧工艺被投入使用时，滤池开始具有生物活性。2017年1月，再对运行进行了调整后，有机物的去除得到了优化。

最初，滤池较低的单位过滤通量（FRVs）被归因于炎热的夏季温度、水厂工作人员还不熟练于新的操作和维护程序，以及对生物滤池通量下降的预估偏少。然而，当2017年夏天，滤池的单位过滤通量（FRVs）继续下降时（图1），这个问题变成了一个重要的关注点。

与切换为生物活性炭过滤工艺之前的一个正常运行的夏季条件下作对比，滤池的单位过滤通量（FRVs）至少下降了30,000 加仑/ft2；然而，即便水厂尽了最大的努力，单位过滤通量（FRVs）也只能做到下降了10,000 加仑/ft2。虽然通量会发生一定的下降是在预料之中的，但现实中这种下降的幅度是出乎意料的，导致水厂已经不可能达到其19.3百万加仑每天的设计生产能力。

Based on the understanding that increased biological growth during warmer weather likely led to the shortened FRVs, plant staff evaluated the biofilters. This was a logical basis for the initial evaluation. Although biological growth has been used in the United States in wastewater treatment for years, its use in water treatment has increased because of the DBPRs. Many water treatment facilities are experiencing a learning curve when it comes to these systems.

In 2018, AWWA’s international symposium on biological treatment extensively covered reduced filter run times and media clogging problems. Discussion with seven other surface water treatment plants that used ozone and biological filtration indicated that approximately 25 percent of plants experienced moderate negative effects, including shortened filter run times, after converting from conventional to biologically active filtration and ozone.

出于对“天气变暖时微生物的生长可能导致滤池过滤通量的下降”规律的考虑，水厂工作人员对生物滤池进行了评估，这也是进行初步评估的逻辑基础。虽然生物法在美国的污水处理中已经应用了很多年，但由于消毒副产物问题的出现，如今在自来水处理中的使用也已经越来越多。当开始真正涉及到这些生物系统时，许多供水企业正在经历一个曲线学习的过程。

2018年，美国自来水协会关于生物处理的国际研讨会的议题中大量涉及过滤时间变短和滤料介质堵塞的问题。通过与其他七个也已应用臭氧-生物活性炭过滤工艺的地表水厂的讨论表明，其中大约25%的水厂在从传统过滤切换为臭氧-生物活性炭过滤后，都经历了一定程度但可接受的负面影响，包括过滤时间变短。

A systematic evaluation followed that investigated each biofilter component, including the following:

■ Assessment of filter underdrain media caps, including evaluation of biology and capacity testing. The PWTF filters use dual parallel later underdrains with media retention caps. Backwashing protocol was also reviewed for conformance with similar facilities.

■ Assessment of the biological growth on the filter media and head loss through the media, including biology and GAC media condition. GAC media condition review included abrasion number,apparent density, ash content, iodine number, and particle size distribution.

■ Evaluation of appropriate ozone dose, backwash procedures, backwash chlorination, influent chlorination, and biomass reduction using a caustic soaking procedure.

随后，该水厂进行了系统性的评估，调查了生物滤池的每个组成部分，包括以下内容：

■评估了滤池的排水管介质盖，包括开展了生物量测试。Poughkeepsies水厂的滤池设计采用了双平行的排水管道，并都安装有滤料的截留盖。工作人员还审查了反冲洗的程序，以便确保与其他水厂类似的设施保持一致。

■评估了滤料介质上的生物生长和通过介质的水头损失，包括颗粒活性炭和生物的状况。颗粒活性炭介质的审查包括磨损数、表观密度、灰分含量、碘值和颗粒大小分布。

■评估了适合的臭氧浓度、反冲洗程序、反洗加氯、进水加氯，以及氢氧化钠浸泡减少生物量的程序。

The evaluation resulted in the following biofilter optimization recommendations that can be applied to any plant with a similar treatment process:

■ Maintaining the recommended ozone-to–total organic carbon (TOC) ratio of 0.5 is important for limiting biomass growth, maintaining plant system redundancy, and conserving energy. Continued observation of ozone dosage is recommended to limit dose while achieving DBP removal potential.

■ The plant should monitor the biomass concentration approximately 6 in. below the filter surface monthly for one year to understand how it varies with temperature and seasonal changes in source water quality. If biomass concentration exceeds 1×10 ⁷  colonyforming units per gram (CFU/g), chlorine at a dose of 0.5 mg/L should be added to filter influent.

■ GAC media has a limited life span. GAC testing is recommended if FRVs begin to decrease.

■ Pressure gauges on the backwash line allow operators to monitor head loss during backwashing. If the pressure required to deliver the same flow rate increases over time, it might indicate fouling is occurring in the filter media and/or on the media retainer cap or sand is penetrating the cap. Excessive pressure can damage the underdrains, so monitoring pressure can help protect the underdrains.

■ Differential pressure meters measuring head loss through the filters should be used to collect additional information when determining whether to terminate filter runs. For example, if head losses of 5–6 ft can’t be reached at the PWTF with the filter effluent valve opened at 70 percent, plant staff should investigate other head loss sources The plant should continue to monitor cleanbed and end-of-run head loss and begin monitoring backwash pressure.

■ Use a different backwash protocol during the summer and winter, as water density is temperature-dependent. Typically, filters are backwashed longer and at a lower rate during the winter. Consider seasonally modified backwash protocols.

The evaluation also determined backwash procedures were effective and there was no filter underdrain clogging, media clogging, or media degradation. None of the usual suspects were contributing to the plant’s low FRVs.

评估工作得出了以下针对生物滤池的优化建议，可适用于任何具有类似处理工艺的水厂：

■ 保持臭氧浓度与总有机碳（TOC）浓度之比为0.5，这对于限制生物量的增长、保持水厂处理系统的冗余度和能源节约非常重要。建议继续观察臭氧浓度变化，并控制臭氧浓度足以实现限值消毒副产物的生成势即可。

■ 水厂应在一年的时间内每月监测滤池表面以下约6英寸处的生物量，了解它如何随温度和原水水质的季节性变化而变化。如果生物量超过每克1×10 ⁷ 个菌落单位（CFU/g），应在滤池进水中加入浓度为0.5毫克/升的氯。

■ 颗粒活性炭介质的寿命是有限的。如果单位通量开始下降，建议需要对颗粒活性炭进行进行测试。

■ 反冲洗管道上的压力表使操作人员能够监测到反冲洗过程中的水头损失。如果保持相同的流速而所需的压力随着时间的推移而增加，这可能表明过滤介质和/或介质固定盖上出现了污垢，或者介质进入了固定盖。过大的压力会损坏反冲洗排水管道，所以监测压力有助于保护管道。

■ 在夏季和冬季使用不同的反冲洗方案，因为水的密度与温度有关。通常情况下，滤池在冬季的反冲洗时间更长，速度更低。应该考虑按不同季节调整反冲洗程序。

评估同时确定反冲洗程序是有效的，之后就再也没有出现滤池冲洗排水管道堵塞、介质堵塞或介质退化的情况。这些通常意义上的“嫌疑人”都没有再引起该水厂的低通量。

While reviewing the backwash procedures, operators observed the 24-in. filter-to-waste piping could pass 3.2 mgd, but the 30-in. filter effluent piping could only pass 1 mgd. This indicated a hydraulic restriction. Although this seems like an obvious red flag, operational procedures were based on flow control valve position, not head loss through the filters. Additionally, head loss gauges measured differential before and after the filter and didn’t reflect a loss occurring after the filter in the effluent piping.

During the backwash procedures, operators observed a tremendous amount of air being released through the filter-to-waste piping. It was theorized that air binding may have caused the issue, thus air release valves were installed on the filter effluent piping. Unfortunately, although the valves effectively released trapped air, they didn’t affect FRVs.

在审查反冲洗程序时，操作人员发现24英寸直径的滤池反冲洗排放管道的流量是3.2毫克/天，但另一根30英寸直径的管道的流量却只有1毫克/天。这表明存在水力通过上的阻碍。虽然这显然是一个明显的故障“红旗”事件，但工艺操作程序的设定是基于流量控制阀的位置，而不是通过滤池的水头损失。此外，水头损失计算的是滤池前后的差值，也并没有反映出滤池排水管道的水头损失。

在反冲洗过程中，操作人员观察到大量的空气通过反冲洗排水管道被释放出来。据推测，管道中存在空气造成堵塞可能是造成这一问题的原因，因此在滤池的排水管道上安装了空气释放阀。不幸的是，尽管这些阀门有效地释放了被困的空气，但它们并没有帮助滤池通量的恢复。

There was only one component left to investigate: the final effluent valve for all the filters. Operators indicated it was opened and couldn’t be closed. By draining the piping and opening the chemical injection tee, the operators were able to insert a camera into the piping and found a large crystalline deposit blocking the valve.

Plant staff removed the valve and replaced it with a spool piece. As shown in the photo at right, the scale blocked most of the valve, severely restricting flow and increasing the system’s head loss.

至此，只剩下一个部件还需要调查：就是滤池最末端的排水阀。操作人员表示它已经被打开，而且也不可能被关闭。排空管道，打开与化学药剂添加管道连接的三通，操作人员将摄像机插入管道，发现是一个大的结晶沉积物堵塞了阀门。

水厂工作人员拆除了阀门，并用一个转轴片代替了它。如右图所示，沉积物堵塞了阀门的大部分，严重限制了流量，增加了系统的水头损失。

A deposit sample was analyzed to determine the scale’s elemental composition. The following parameters were quantified: aluminum, calcium, iron, magnesium, manganese, sodium, phosphorus, and TOC. Calcium was the most abundant of the analyzed elements, but a substantial fraction of the sample was unknown. Inorganic carbon (i.e., carbonate) might have contributed to the unknown fraction.

为了确定水垢的元素组成，工作人员对沉积物样本进行了分析。对以下的成分物质进行了量化：铝、钙、铁、镁、锰、钠、磷和TOC，其中钙是所分析到的元素中最丰富的，但也有相当一部分物质是未知的。无机碳（即碳酸盐）可能对未知部分有贡献。

The valve is located immediately downstream of sodium hypochlorite (NaOCl) and sodium hydroxide (NaOH) injection points. Both chemicals are added continuously, but the NaOH dose was increased by a factor of approximately 3 (from a dose of 3–5 mg/L to 13.5–17.5 mg/L) when enhanced coagulation was practiced to remove organics before the ozone system became operational. The plant practiced enhanced coagulation with carbon dioxide addition prior to solids contact clarifiers from January to August 2017, which was the same period when the filters became biologically active. Thus, the deposit caused the low FRVs, but it was obscured by the simultaneous conversion to biofiltration.

该被堵塞的阀门位于次氯酸钠（NaOCl）和氢氧化钠（NaOH）药剂投加点的下游。这两种化学品都是连续添加的，但在臭氧系统开始运行之前，为了去除有机物而进行强化混凝时，NaOH的剂量增加了大约3倍（从3-5mg/L的剂量增加到13.5-17.5mg/L）。该水厂在2017年1月至8月期间，在接触澄清器之前实施了添加二氧化碳的强化混凝，这也是滤池培养维生物使具有生物活性的同一时期。因此，就是在这个时间段内，沉积物导致了滤池的低通量，但真正的原因却被“误导”为了生物过滤池的原因。

It’s likely this increased NaOH addition caused a locally elevated pH condition, which led to calcium precipitation and scale formation. The calcium concentration in the filter effluent is relatively low at 22–36 mg/L, but as shown in the calcium solubility diagram in Figure 2, it’s firmly in the solid phase (indicated in blue shading) at this concentration (10 ^–3 to 10 ^–4 M) once pH reaches 10. It’s likely a high pH occurred near the NaOH injection point because the NaOH lacked sufficient time and turbulence to mix completely.

分析原因，很可能是因为增加的NaOH投加量造成了局部pH值升高的状况，从而导致了钙的沉淀和水垢的形成。滤池出水中的钙浓度相对较低，为22-36毫克/升，但如图2中的钙溶解度图所示，一旦pH值达到10，且钙元素在10 ^-3 到10 ^-4  log浓度这个浓度下，就会“牢固”地处于固相的区域（用蓝色阴影表示）。结垢的情况很可能就发生在NaOH注入点附近，因为NaOH缺乏足够的时间和湍流来完全混合，造成了局部的高pH值。

Assuming only 1 percent of the calcium in the filter effluent contacted the high pH area and was converted into solid calcium carbonate (CaCO3), a precipitation rate of 32 in.3/mil gal of filter effluent could occur. For comparison, a tennis ball has a volume of 8 in.3 Not all precipitated CaCO3 would be expected to scale the filter effluent valve, but this growth easily could have occurred over a relatively short period of time.

假设滤池的出水中只有1%的钙接触到上述的高pH值区域，并被转化为固体碳酸钙（CaCO ₃ ），那么滤池出水中就可能会出现沉淀率为32立方英寸/密耳加仑（1密耳=0.0254毫米），形象地对比一下，一个网球的体积为8立方英寸。并不是所有沉淀的CaCO ₃ 都会在滤池的出水阀上形成水垢，但水垢的形成很容易在相对较短的时间内发生。

Well-functioning biologically active filters are critical to helping PWTF meet its hydraulic capacity and water quality targets. Biofilters help PWTF meet its treatment objectives by lowering DBP formation potentials to meet the DBPRs and lowering the assimilable organic carbon concentration, which helps limit DBP formation and microbial regrowth in the distribution system.

Although interviews with other biofiltration plants indicated biofilters can experience seasonal shortened filter run times, the biomass in the filters was determined not to cause the shortened FRVs experienced at PWTF. They were caused by the formation of a CaCO3 scale on the filter effluent valve. Once the valve was removed and replaced by a spool piece, FRVs immediately increased (Figure 3). In the summer of 2019, FRVs were typically between 20,000 and 25,000 gpd/ft2, indicating an FRV reduction consistent with expectations.

运行良好的生物活性炭滤池对帮助水厂达到其产水水量和水质目标至关重要。生物滤池能够帮助Poughkeepsies水厂实现其水处理目标，降低消毒副产物生成的可能性，以满足美国环保总署的消毒副产物控制规则，并也可以降低可同化有机碳的浓度，这将有助于限制输配系统中消毒副产物的形成和微生物的再生长。

尽管对其他生物活性炭工艺的水厂的采访表明，生物滤池可以随着季节性的变化缩短过滤运行时间，但滤池中的生物量也被确定为并不是Poughkeepsies水厂导致滤池通量下降的原因，通量的下降是由滤池出水阀上形成的CaCO ₃ 水垢造成的。被堵塞的阀门一经被移除，通量立即被恢复（图3）。在2019年的夏天，滤池的通量通常在20,000到25,000 加仑/每天每平方英尺之间，说明滤池通量的变化与预期一致。