Achieving the perfect mix

A greater volume and wider variety of sludge than ever is being processed in the UK. Mixing is a fundamental operation in sludge processing, yet it is still considered by many to be a black art.

Anaerobic digester mixing is vital in order to blend digester feed sludge, make use of the whole digester volume, dilute inhibitory substances, avoid temperature or concentration stratification and maximise gas production. The increased emphasis on pathogen destruction in digesters has dictated that retention times in the digester must be optimised by avoiding dead zones and short-circuiting. There is also a need to digest thicker sludges in order to increase throughput.

The effective operation of sludge thickening and dewatering processes is dependent on the upstream addition of sludge conditioning chemicals such as polyelectrolytes. The rate of mixing of the polyelectrolyte can have a big influence on the thickening or dewatering performance as well as the dose of polyelectrolyte needed. Getting polyelectrolyte mixing right can therefore improve dewatered sludge dry solids, reducing the cost of transport and/or further processing. For example, on one belt-press plant, intensifying polyelectrolyte mixing improved polymer effectiveness (kg dry solids/kg polymer)

by 40%.

In addition to the common sludge mixing applications described above, there are plenty of novel processes being adopted by the industry – thermal and enzymic hydrolysis,

various chemical treatments and pasteurisation – which require mixing steps. Some of these processes involve sludges with distinctive properties that have not been encountered before, making the task of mixer design or selection very difficult.

Performance gauged

In order to get the best from a process we must quantify the mixing performance and compare what we have with the mixing performance we think we need. To do this we can look at the flowpattern, residence time and mixing rate. The flowpattern describes the way sludge flows through the tank or mixer, the two ideal (extreme) flowpatterns are plug-flow and fully back-mixed. Most processes should operate at one of these ideals but in practice non-ideal short-circuiting, stagnant regions or recirculation loops are present. The flowpattern can be quantified by measuring or modelling the distribution of sludge age as it flows through the tank or mixer. This will give the residence time distribution.

Most sludge mixing applications involve blending or homogenisation. To quantify this we need to determine mixture quality (concentration variation) and time taken to reach it (blend time or mixing time). Mixture quality is described using the coefficient of variation (CoV) which is a statistical measure of concentration variability. The smaller the CoV value, the better the mixture quality. A CoV of 0.05 is a frequently used

target value, this roughly equates to measured concentration varying within +/- 10% of the mean value.

Two key questions need answering for specification and design of sludge mixers. How good does the mixing need to be to optimise the process? How can you achieve that degree of mixing?

The answer to the first question requires consideration of the mixing objectives. This should always be carried out before mixing equipment is sized or selected. Digester mixing systems, for example, might aim to:

• achieve an actively mixed sludge volume of 90%,

• ensure the short circuit volume is less than 5%,
• blend feed sludge into the bulk within two hours.

Polyelectrolyte conditioners dosed into sludge prior to dewatering may need to be blended to CoV ≤ 0.05 within one or two seconds. This rapid mix should be followed by a period of less intensive shear for floc conditioning. Sludge storage tanks may need to be mixed to CoV ≤ 0.05 prior to emptying.

It is important to realise that the optimum degree of mixing, especially of dosed chemicals, may be specific to your process and hence may need to be determined by jar tests or similar scaled-down procedures.

Once quantified mixing objectives have been decided and other constraints identified, the most cost-effective mixing equipment selection can be made. Not all mixer types/designs require the same amount of energy to achieve a specific mixing objective, so it pays to understand the capabilities and relative performances of the available kit. BHR Group is an independent centre of mixing know-how and has an ongoing programme of research into sludge mixer performance called WWM. The company also provides mixing and fluid engineering consultancy services.

There is a large selection of equipment available for sludge mixing. Tank or digester systems include numerous types of gas injection, pumped liquid jets, top-mounted impellers and submersible mixers. There can be major differences in energy consumption for these systems. Research at BHR, sponsored by Yorkshire Water and Monsal, has shown that well designed sequential gas mixing systems can significantly outperform those where gas is supplied continuously to multiple nozzles. For example, replacement of continuous mixing systems with sequential ones realised a power saving of £3,950 pa per digester at Mogden WwTW.

In-line sludge and conditioner mixing can be achieved using T-mixers, static mixers, valves, pumps and dynamic (rotating) mixers. The mixing efficiency and performance of these devices varies dramatically, as does their tendency to block. However, little or no published information on the relative performances of this important class of equipment is available, a situation the WWM project hopes to rectify.

It is vital to know how your sludge flows (sludge rheology) when sizing or selecting a mixer for sludge applications. Sludges often exhibit non-Newtonian shear-thinning rheology where viscosity varies with shear-rate and hence position within the tank or mixer. In addition, some sludges exhibit a yield stress which must be exceeded if the sludge is to flow at all. Sludge rheology is difficult to predict on the basis of dry solids, temperature and sludge type alone, this is particularly so for sludges associated with newer processes. Sludge rheology should therefore be measured to enable sludge process equipment to be designed properly.

The performance of in-line sludge mixers is drastically influenced by whether the prevailing flow regime is laminar or turbulent, the latter is much faster. The flow regime is in turn strongly influenced by sludge rheology. Without an accurate knowledge of the sludge rheology you cannot determine mixing performance and realise potential savings.

Making Predictions

Mixing performance can be measured using physical or computational model studies as well as on-site tracer testing. Mixing research for the chemical process and water industries has relied on model studies at different scales to identify the influence of different variables and to build-up predictive correlations for mixing performance. On-site tests can be used as a diagnostic tool and to validate the performance of installed equipment. For example, LiCl washout tracer tests can be used to measure digester active volume and blend time.

Attention to sludge mixing pays. The first step is to define the mixing objectives for your sludge process application. Next it is important to measure the flow behaviour of your sludge.

Once this has been done, you can select the most efficient mixers to achieve your mixing objectives. Ideally this should be based on independent information, alternatively mixer suppliers can be consulted. Once the chosen mixers are installed the mixing performance

relative to your objectives should be measured on-site to confirm the suitability of the installed equipment.

There is considerable scope for reduction in sludge processing costs through improved understanding of the link between mixing and process performance. In addition, better knowledge of relative mixer performance for sludge applications will lead to optimised processes and the emergence of more suitable mixer designs.