‘Technology transfer-oriented research and development in the wastewater sector – validation at industrial-scale plants’ (EXPOVAL) – Subgroup 6: solar sewage sludge drying: first results from investigations with a pilot plant

The requirements for wastewater treatment worldwide are becoming more complex. In addition, the sewage sludge produced needs to be treated adequately to meet the sludge reuse and/or disposal requirements. Design rules for wastewater and sludge treatment have been established in Germany. To apply these in foreign countries, adaptations are necessary, especially in relation to local climatic conditions. One of the most important treatment steps for reducing sludge mass is removing water from the sludge during drying. One suitable technology for stabilized sewage sludge is solar drying. It is relatively easy to implement and to handle, and operating costs for electricity and heat are low. In addition, the pathogen content in the sludge solids can be reduced by solar UV radiation while much of the water is removed. The sludge solids produced using this technology are suitable for agricultural reuse. Many solar sludge drying plants are already in use. Experience in constructing and operating them is available, but design rules, especially for different climatic regions, does not exist. Hence, the general objective of this project is to determine and quantify factors influencing sludge water evaporation, so that it can be predicted on the basis of climatic and operating data. Currently, several investigations are in progress regarding different external heat input devices so that sludge drying with supplementary heat is also possible in temperate zones in winter. The additional energy demand for external heat input must be balanced against the advantages of improved total drying performance and the smaller drying area needed.

The investigations of solar sludge drying are part of a joint research project called EXPOVAL, financed by the German Federal Ministry of Education and Research. The project objective is full-scale validation of municipal wastewater and sludge treatment concepts under varying climatic conditions, looking at those which appear to be especially promising in regard to treatment results and economic efficiency. Existing German design rules will be adapted and extended, and new rules developed where none exist at present. The boundary conditions include a wastewater temperature range between 5 and 30 °C, and salt (NaCl) contents exceeding 2 g/L. Because of this, investigations of commonly used technologies are being carried out on all continents in various climatic zones (Emscher Wassertechnik GmbH 2012). The joint project is subdivided into seven groups working on different wastewater and sludge treatment technologies, so that recommendations on design and operation for complete treatment chains can be presented. The results will be published in a technical guide by the German Association for Water, Wastewater and Waste.

Subgroup 6 of the joint project is validating existing, full-scale solar dryers for sludge in different climatic regions and developing design rules for the technology. For this purpose, a pilot-scale solar dryer has been constructed by HUBER SE. This pilot unit is operated in parallel with full-scale solar dryers, to quantify the different factors influencing the drying process and to optimize operation. Investigations in Penzing and Braunschweig, Germany, were used basically for optimizing and upscaling the pilot unit. On-site investigations will also be carried out at Cali in Columbia (subtropical climate), and in Klodzko, Poland, with a temperate climate and cold winters.

The first investigations started in 2013 at a full-scale solar dryer in Penzing, in order to adjust the pilot plant operation to full-scale performance. In addition, the pilot-scale unit was placed in a building in Braunschweig, so that the influence of climatic factors could be minimized and the full effect of operational adjustments on drying performance could be considered.

Any water remaining in sewage sludge after dewatering can be removed only by evaporation or volatilization. It cannot be removed mechanically and thermal energy is needed to achieve a higher total solids (TS) content. The TS produced can be reused in agriculture, or incinerated or landfilled. Thermal sludge drying technologies rely on one or more of three physical processes – namely convection, conduction and radiation – for heat transfer.

In convection systems, the wet sludge comes into direct contact with the heat-transfer medium, usually hot air. The air circulates round the sludge and heats the particles; the water is transferred from the sludge to the gas. In conduction systems, heat is transferred to the sludge through a solid barrier. Steam or thermo-oil is usually used as the transfer medium and does not come into contact with the sludge. The vapor from the sludge is removed separately from the heat transfer medium. In radiation drying, the energy is transferred without a physical heat carrier, by electromagnetic or infrared radiation. Heat is created within the sludge when electromagnetic radiation is converted into thermal energy (DWA 2005).

Solar drying is a combination of convection and radiation effects. Sewage sludge that has been dewatered mechanically is spread on a concrete floor in a transparent drying hall (similar to a conventional greenhouse), providing an enhancement on open air drying beds without climatic influence. As the sludge is protected from rain, the drying area and time are lower than in open air drying beds. External heat input can enhance the drying process, e.g., by under floor heating integrated into the bottom plate or by air circulation. The walls and roof of the drying hall are transparent, e.g., glass, Plexiglas, polycarbonate or polyethylene (Bux 2013), and create a greenhouse effect. Both visible light and shortwave infrared radiation pass the transparent material, and radiation striking the sludge surface is reflected as long-wave thermal radiation that cannot leave by the same route. Thus, the energy stays in the drying hall and warms the internal air, which can then absorb more moisture. When cold air enters the dryer it, too, is warmed and can take up additional water from the sludge. Thus, the evaporation rate is increased significantly and the drying process accelerates. Regular agitation breaks up the dried surface of the sludge and increases evaporation from beneath (Lue-Hing et al. 1992). To avoid the formation of anaerobic zones, the sludge is mixed and turned frequently. Solar dryers can work in batch as well as continuous mode. There are nearly 300 such plants in operation in Europe (Jacobs 2013) with many more elsewhere, and they treat sewage sludge from population equivalents of between 1,000 and 600,000 (Bux 2013).

Throughout the drying process, samples were taken regularly from both the pilot- and full-scale plants, using a cylinder to take cores. Sampling points were located regularly across the drying area, and the samples were analyzed for a variety of parameters using the methods shown in Table 1.

A pilot plant was built to assist in developing design rules and validating the existing solar dryers. The unit is mobile and built at reduced scale – see Figure 1. It consists of a drying area of about 7 m² and a transparent cover made of UV-resistant polycarbonate. It is sized at a scale of 1:100 compared to full-scale plants and fits into a 20-ft intermodal container, to make it easily transportable. The drying area is a sludge pan separated from the rest of the plant and can hold up to 1,000 kg of dewatered sewage sludge. Changes in sludge masses are weighed and logged automatically. The sludge is mixed and turned by a rotation device on a rail over the sludge bed. The mixing frequency can be adjusted manually over a wide range. Two fans keep the internal air turbulent, and moisture-saturated air is removed and replaced. The ventilation fans can be replaced by a fan heater and under floor heating beneath the sludge bed can be used to investigate other heat input devices. Unlike full-scale plants, the pilot plant was constructed for batch tests, but semi-continuous processing is possible as well.

The pilot plant includes devices for measuring sludge weight, surface temperature, ammonia emissions, sludge layer level and solar radiation intensity, as well as temperature and humidity both inside and outside the dryer. The energy consumption of all electrical and mechanical devices can also be recorded.

Initial investigations, focusing on upscaling of the pilot-scale unit were carried out in Penzing. This site was chosen because it was close to the pilot unit's manufacturer and the existence of a full-scale plant there. The pilot plant was placed close to the large dryer and facing south, so that most of the sludge surface could be reached by the sun.

After the work in Penzing, the pilot unit was operated in a building in Braunschweig, to assess the influence of external heat input on drying, energy demand and odor generation with weather influences minimized. Under floor heating and a fan heater were used separately as external heat sources. To characterize the drying process, the TS content of the sludge, energy consumption, climatic data and ammonia concentrations in the plant were measured regularly. Six batch operations were run under different operating conditions, and the energy input conditions for the different batches are given in Table 2.

In all experiments, the same digested and dewatered sewage sludge from a municipal waste water treatment plant was used. Dewatered sludge with 26% TS was added once during drying in batch 4 and twice during batch 5 to get an impression of the consequences of a semi-continuous operation.

Because the pilot unit was placed in a building, weather influences were minimized. Neither precipitation nor radiation affected the drying process, although small variations in temperature and humidity were detected through the daily cycle. In Penzing relative humidity varied between 20 and 100% within a day. In Braunschweig, with reduced atmospheric influences, relative humidity varied about 10% within 24 hours. During one drying period relative humidity ranged between 30 and 90%. Temperatures within the pilot plant varied within five Kelvin during the day in Braunschweig, compared to diurnal differences of up to 30 K during operations in Penzing. The temperature and relative humidity of batch 4 are shown as an example in Figure 2. In the following explanations the term ‘inside’ means ‘within the pilot plant’, while ‘outside’ means ‘external to the plant but within the building’.

Figure 2 shows that the inside temperature was slightly higher than that outside and that both varied during the day. The inside temperature varied little during the drying process, until toward the end of the batch trial when the TS reached 80%. On the other hand, relative humidity inside the plant was much more variable. It reflects the conditions outside, but has much more local maxima and minima, resulting from fan aeration. As the TS rose to levels above 80%, humidity decreased rapidly, approaching the levels found outside. This indicates that most of the water had evaporated from the sludge and no additional water was taken up by the air. The external heat input from under floor and fan heating could be adjusted: one Kelvin steps for the under floor heating and three different levels for fan heating.

During the 4-month operation in Penzing, evaporation rates between 105 and 344 g m^-² h^-1 could be calculated for the pilot-scale plant. These rates agree reasonably well with data in the literature, where values are between 79.9 and 137 g m^-² h^-1 (Kassner 2000 converted). Comparing those values, it must be noted that the investigations in Penzing were scheduled from April to July (4 months with naturally high evaporation), whereas Kassner's numbers result from annual monitoring, including wintertime, leading to lower mean values.

Analysis of the evaporation rate in the full-scale dryer was more difficult than in the pilot plant, as fewer process data are recorded and the operation is variable. Apart from that, the total sludge mass could not be measured during the drying process, so that mass balances had to be calculated based solely on input and output masses. The back-mixing of dry sludge into the wet sludge led to further inaccuracies. Differences in evaporation rates from the pilot- and full-scale plant could be observed, especially in a sequential pair of batch trials in June and July 2013. Nevertheless, for the full-scale operation evaporation rates of 555 (batch trial 1) and 154 (batch trial 2) g m⁻² h⁻¹ could be calculated. Simultaneously, in the pilot plant, evaporation rates of 344 (batch trial 1) and 177 (batch trial 2) g m^-² h⁻¹ were measured.

It is possible that the different evaporation rates in the pilot- and full-scale plants arose from different aeration, operational aspects and sludge feeds. The aeration differed in timing and volume because different aeration mechanisms were involved: in the pilot plant, discharge of moisture-laden air was controlled by a fan, with fresh air replacing it automatically. In the full-scale plant, air inlet and outlet were controlled by opening and closing windows, with fans moving the air through the drying hall. Daily variations in operation in the full-scale plant also need to be considered. The plant operators adjusted the frequency and speed of sludge mixing and feeding to the dryer in relation to the weather conditions, the amount of sludge and staff capabilities.

As the size of the dried sludge granules is another aspect of interest for upscaling, screen analyses of dried sludge solids from the pilot- and full-scale plant were carried out. Four sieve sizes were used – see Figure 3. As can be seen, both grain-size distribution curves, from the full- and pilot-scale plant, show similar size fractioning.

No granules were bigger than 16 mm, and nearly 50% of the mass could be found in 5 and 0.5 mm sieves, while more than 90% of the granules were between 0.5 and 10 mm. The fraction of particles smaller than 0.5 mm is only about 1%. In general, the results show that the pilot- and full-scale plants were comparable with respect to their dried solids outputs.

The investigations showed that sludge drying in the pilot plant inside the building and using external heat input was feasible. Evaporation rates were of the same order as those achieved in Penzing. The results shown below focus on batch trials 3 and 4, the boundary conditions for which were almost the same. The same sludge was used and the mean outdoor temperatures were 19.5 °C and 16.8 °C, respectively. Figures 4 and 5 show the sludge weight, dry solids content, and ammonia concentrations as they changed during the two trials.

As can be seen in Figures 4 and 5, continuous processing could be carried out using external heat input, and there is no apparent significant difference between the energy sources concerning weight loss. Classical drying curves (e.g., Loll & Melsa 1995) could be replicated with a fast decline in weight at the start and slower at the end. At the same time, the TS rises gradually from the 27 or 28% of the dewatered digested sludge to over 90%. The TS content was measured daily and the resulting curve fits quite well with that of weight loss, which was measured automatically four times per hour. The correlation of weight reduction and TS increase due to evaporation is clear from the curves. In the fourth batch trial (Figure 5), 75 kg of fresh dewatered sludge (26% TS) were added after 9 days of processing when the original sludge mass was reduced by 68% reaching 91% TS, to evaluate the effects of semi-continuous operation.

To compare the two different methods of energy input, batch trials 3 and 4 were considered between 30 and 89% TS content. For batch 3, the fan heater was used at level 2, and regulated by the inside air temperature, leading to surface temperatures between 25 and 35 °C. Energy input for batch 4 was achieved using the under floor heating, controlled at 55 °C, leading to surface temperatures between 25 and 30 °C. Table 3 shows the differences in drying behavior throughout one drying process.

Although batch 3's total drying time is 10 hours longer than batch 4's, the load in batch 4 was around 50 kg higher, so the drying speed to reach 89% TS was also greater in the fourth batch: the mean evaporation rate reached 254 g m^-² h^-1 compared to 202 g m^-² h^-1 for batch 3. This shows the advantage of under floor heating compared to a fan heater.

The enhanced evaporation using under floor heating arises from the heat transfer mechanisms. Thermal energy is supplied directly to the sludge and passes to the air, so it is used directly to transfer water to the air. The energy from the fan heater, on the other hand, must pass through the air before water can be evaporated and discharged.

The faster drying rate arising from under floor heating also appeared to have a lower energy demand. This is primarily because it is relatively independent of air transmission compared to energy from the fan heater and the temperature is controlled directly with under floor heating. The fan heater, on the other hand, is regulated according to the inside air temperature, and switched on and off frequently. The sporadic operation of the fan also led to the continuous discharge of warm air not fully laden with water. Adjusting the ventilation to achieve lower frequencies could improve drying.

The ammonia concentration in the air discharged from the plant was measured as an indicator of odor emissions. Concentrations during drying correlated with the drying progress of the sludge. In the beginning, a large amount (exceeding the measurement limit of 60 ppm) of ammonia was released, regardless of the heat source or its temperature level. Atmospheric ammonia concentrations decreased sharply above 35% TS, so that they were more or less constant at around 20 ppm for several days and then approached zero. Only when there was no further change in sludge weight, did ammonia emission concentrations remain constant. Whenever new dewatered sludge was added to the plant (see Figure 5) a sharp rise in atmospheric ammonia concentration was observed, although the level was much lower than at the beginning of the trial. Thus, if the plant is run in semi-continuous mode, the peak emissions are lower, although the total amount of ammonia discharged is the same – regular doses of relatively low concentration are discharged instead of a single, highly concentrated release. Because of this, semi-continuous operation would enable, say, German Technical Instructions on Air Quality Control (BMU 2002) emission standards to be met.

Exemplary measurements of Kjeldahl nitrogen in the sludge showed a reduction of 53% during drying. Sludge samples were taken in the beginning and at the end of drying, to determine nitrogen compounds, and analyzed for ammonium and TKN. The dewatered sludge contained 8.83 kg TKN of which 4.15 kg remained in the dried material, i.e., more than 50% of the ammonium had been transferred to the gaseous phase.

Pathogen analyses showed a reduction of indicator bacteria at both 55 and 80 °C floor heating temperatures. Salmonella could not be found in any of the dried sludges. The fecal coliform content of the sludge dried at 55 °C was reduced from 1.39 × 10⁴ to 4.35 × 10³ MPN/g TS. The sludge dried at 80 °C showed a reduction from 2.77 × 10⁶ to 1.39 × 10⁴ MPN/g TS. These data indicate that partial disinfection can be achieved. For adequate disinfection, i.e., fecal coliforms ≤ 1,000 MPN/g TS, as required by some standards (e.g., USEPA 2007), higher floor temperatures and adjustments in operation might be necessary.

Initial investigations with the pilot- and the full-scale plant show, that the drying performance of the pilot plant can be adjusted to match that of full-scale operation. This is the key to further development and enhancement of the design and operation recommendations, especially in relation to different climates. Here, especially the fractioning of the dried sludge as measured in screen analyses showed comparable results.

Under floor heating appeared to yield evaporation rates that were 25% higher and yet have a lower energy demand than alternatives studied at the same time. Both under floor heating and fan heater systems achieved evaporation rates comparable to full-scale solar drying under German spring/summer conditions. Ammonia emissions correlated with the drying progress, the high emission peaks during initial feeding of dewatered sludge can be reduced by semi-continuous operation.

After the batch tests in Germany, further investigations will be carried out in Cali, Colombia. At WWTP Cali, the pilot plant will be set up next to a full-scale solar dryer. The main focus of these investigations will be on the influence of different operational aspects and on upscaling effects in regard to evaporation rate calculations in relatively constant subtropical climatic conditions. Afterward, the pilot plant will be transferred to Poland, where investigations will focus on external energy input for sludge drying during wintertime.

The project 02WA1252P/02WA1252Q is financially supported by the German Federal Ministry of Education and Research (www.bmbf.de; www.expoval.de/en).