SEM pictures of the skin and support layer of the unused membrane showed that the grain size of the support material was much bigger than that of the membrane skin layer. The skin layer of the studied membrane was outside of the tube. On the support layer some particles, which probably are dust, could be seen on the top of the bigger membrane particles. From the cross-section pictures, the thickness of the membrane layers was determined. The support layer was approximately 1.5 cm thick and the skin layer 63 mm.

The SEM pictures of the membrane treated with hydrogen peroxide (40 vol.-%, 19 h) and phosphoric acid (80 %, 24h) showed that in phosphoric acid treated membrane some particles were etched more together than in peroxide treated membrane. It also seemed that the phosphoric acid treatment eroded the grains, so that the spaces between some of the grains were bigger. When comparing the peroxide treated membrane to the unused membrane, it was obvious that hydrogen peroxide treatment did not have any visible effect on the membrane structure, but that phosphoric acid may have changed the surface structure slightly. However, the changes were not so significant, and it was difficult to say whether they really were caused by the acid or by normal differences in the membrane surface. The exposure time should probably have been longer in order to be able to make more reliable conclusions. The cross-sections of treated membranes did not show any significant visible differences. Moreover, no visible differences were observed when the cross-sections of the treated membranes were compared to that of the untreated membrane.

The cross-section pictures of the membrane used in juice filtration showed a significant difference in the thickness of the skin layer compared to that of the unused membrane. The thickness of the skin layer of the membrane used in the juice filtration was only from 8.0 mm to almost zero. SEM pictures of the skin layer of the membrane used in cherry juice filtration tests showed that in some places the big particles from the supporting structure came through the skin layer. Moreover, the material that formed the skin layer seemed to be much smaller in particle size than the material in the unused membrane.

The first assumption when looking the skin layer of the used membrane was, that the small particles on top of the membrane were foulants. However, this probably was not the case as the cross-section picture showed. There are two possible explanations for the differences in the skin layer: either the manufacture of the membrane skin layer had failed, or the backshock had damaged the skin layer. The first option is the most likely. The fouling layer, if it existed, was difficult to see from the skin layer of the membrane, partly since during the cutting of the membrane pieces, the dust was attached to the surface of the membrane. However, rests from the juice could be observed on the support layer. Grain particles could be distinguished on the support layer. However, the sharpness of the grain borders had vanished. The bigger magnification showed that particles were covered with some kind of a gel layer.

The SEM pictures of the membrane used for the filtration of the cherry juice and then cleaned with both acidic and alkaline cleaning agents showed that the cleaning agents had removed part of the fouling layer. The fouling was more evident in the cleaned than in the used membrane, partly because this membrane had not been cut with a blade and therefore there was no dust attached to the membrane. From the cross-section pictures of the used and cleaned membrane, some clusters of small particles could be seen in between the bigger particles. These small particle clusters could be either foulants or the material which was supposed to form the skin layer, but which instead had filtrated inside the supporting structure.

The water content of the piece of used membrane was measured by weighting this before and after drying. The membrane was dried in the oven at 105ºC for 4 h before been placed in the dessicator for 2 h. The amount of water on the membrane was 0.2201 g. The weight of the dry membrane was 1.6093 g. The membrane piece was stored in the dessicator for 21 days until it was incinerated in the oven at 550ºC for 2 h. The membrane was weighted again before incineration and its weight was 1.6075 g. This means that that 0.0018 g were lost during the storing in the dessicator. During incineration, the following program was used: First, the temperature was increased 5ºC/min to 150ºC and this temperature was held for 1/2 h. Then, the temperature was increased 5ºC/min to 550ºC and then held for 2 h. After that, the temperature decreased 10ºC/min to 20ºC. Membrane pieces (both unused and used) were placed into the oven when the temperature was 130ºC. The used membrane looked yellow when it was taken out of the oven, but after cooling in dessicator, it was white. The unused membrane was white when it was taken out of the oven. In Table 7.3 the results from the drying and the incineration with calculated weight losses are presented.

Table 7.3. The weights of the unused and used membrane before and after drying. Abbreviations: D m weight loss, n.m. not measured.

The water content in the used membrane was 13.7% and the organic content 0.19%, which means that the amount of water was 137.2 mg/g of incinerated membrane and that of organics 1.93 mg/g incinerated membrane. The loss of matter during storing was 1.12 mg/g of incinerated membrane. The amount of total organics in the unused membrane was only 0.13 mg/g of incinerated membrane. The amount of water in the membrane after use was high. However, the total amount of organics in the used membrane was not that high.

7.2.2.3 Protein analysis

One piece of dried and unused membrane, and one piece of dried and used membrane were analysed for amino acid content. The analyses were made according to the following procedure: membrane pieces were hydrolysed in 1 ml 6N HCl, 50 m l phenol and 50 m l DTDPA in a big beaker, and then washed in another beaker with 400 m l 0.1M HCl. After that, the sample pieces were discharged. Both the hydrolysing solution and the washing solution were then dried, and the amino acids were redissolved. The amino acids from the hydrolysing solution were redissolved in 1000 m l of a buffer and those from the washing solution in 100 m l of the same buffer. Then, the amino acids were determined by Liquid Chromatography. The most abundant amino acids in the used membrane were glycine, aspartic acid, proline, serine and glutamic acid, whereas in the unused membrane were glycine, glutamic acid, serine, aspartic acid and histidine. The peak of the TP (thiopropionic) cysteine could be detected clearly in the used membrane, but not in the unused membrane. The total amount of proteins in the different samples is presented in Table 7.4. The total amount of proteins for the used membrane was 217 m g/g membrane and 1.07 m g/g for the unused membrane. The results show that the hydrolysis did remove most of the proteins but not all. Furthermore, a significant amount of proteins was also found. Since washing was done only once there exists a small possibility that some of the proteins stayed in the membrane.

Table 7.4. Total amount of proteins in different samples.

7.2.3 Analysis of phenols

First, dried sample pieces of unused and used membrane were weighed. The weight of the unused membrane piece was 1.5888 g and that of the used membrane 1.6071 g. Then, samples were crushed into small pieces in a mortel in order to increase the surface area for extraction. The powdered sample was put into a glass bottle and the mortel was rinsed four times with approximately 10 ml of 70 vol.-% acetone and poured into the same bottle. Acetone and the powdered sample were well mixed and left over night (at least 14 h) to extract. Before analysis, samples were centrifuged at 10 000 rpm for 10 min in order to separate the membrane particles from the extractant. The supernatant was analysed for total phenols according to the following procedure (Singlenton et al., 1965):

1 ml of Folin-Ciocalteu’s phenol reagent, diluted 1:10 with double distilled water was added to 0.2 ml of phenolic extract. 0.8 ml of 7.5% (w/v) sodium carbonate was added to develop the colour and everything was mixed. The mixture was then left for 30 min with caps on. After mixing again, the absorbance was read at 765 nm, using double distilled water for background correction. The concentration of total phenols was calculated from a standard curve obtained by subjecting known amounts of gallic acid solutions (0, 5, 10, 20, 40, 60, 80 and 100 mg/l gallic acid) to the same treatment as the test samples. Results were expressed in gallic acid equivalents (GAE).

The equation for the standard curve was c(mg/l GAE) = 94.276× ABS-2.2123. The results from the measurement are presented in Table 7.5. The total amount of phenols in the used membrane was 228 m g/g membrane and 72 m g/g in the unused membrane.

7.2.4 Summary and conclusions

In this work an unused, modified ceramic membrane, a used ceramic membrane, and a cleaned ceramic membrane were studied in order to find out some of the membrane foulants and the effect of some cleaning agents on the membrane. The analysis of foulants was done by microscopic means and by analysing some specific groups of substances, which may cause fouling, as well as by determining the total amount of organics on the membranes. The effect of the cleaning agents on the membrane was studied by microscopic means.

Microscopic studies were done by Scanning Electron Microscopy (SEM).The SEM pictures of unused and used membranes showed a significant difference in the thickness of the skin layers. It seems that the manufacture of the skin layer did not succeed in the membranes used in juice filtrations, and that the particles meant to form the skin layer were filtrated inside of the supporting structure. The SEM pictures of the membranes treated with hydrogen peroxide and phosphoric acid did not show visible differences. However, the phosphoric acid treated membrane seemed to have a slightly different surface structure, but it is difficult to conclude whether the differences were caused by the acid or whether they were just part of normal variations of the membrane structure. The pictures taken from the used and the used and cleaned membrane showed that there was fouling on both sides of the membrane. The cleaning could not remove all the foulants, and some of them could be seen after cleaning forming a sponge-like structure. Since the filtration was made from inside to outside (from support layer to skin layer), it is understandable that the foulants could be found at both sides of the membrane.

The amount of water and total organics on the membrane was determined by drying the membrane at 105ºC and then incinerating it at 550ºC. The amount of water in the used membrane was 137 mg/g of incinerated membrane after two hours of evaporation. However, some substances (1.12 mg/g incinerated membrane) were evaporated from the membrane during storing in the dessicator. The amount of total organics in the used membrane was 1.93 mg/g incinerated membrane, whereas in the cleaned membrane this value was 0.13 mg/g incinerated membrane.

Proteins were analyzed from the unused and used membrane by degrading them into amino acids with HCl and analysing the amino acids by Liquid Chromatography. The total amount of proteins in the used membrane was 217 m g/g dry membrane and in unused membrane 1.07 m g/g dry membrane. In the chromatograms of the used membrane the most abundant amino acids were glycine, aspartic acid, proline, serine and glutamic acid, whereas in the unused membrane the most abundant were glycine, glutamic acid, serine, aspartic acid and histidine. The peak of the TP Cysteine could be detected clearly in the used membrane, but not in the unused membrane.

The total amount of phenols was analysed from the unused and used membrane by extracting them with 70 vol.-% acetone (14 h), and then determining them from the extract with a colorimetric method. The total amount of phenols in the used membrane was 228 m g/g dry membrane and in the unused membrane 72 m g/g dry membrane.
A summary of the results achieved from the different analysis is presented in Table 7.6. The amount of foulants in the unused membrane was very small, and those values can be used for the estimation of the accuracy of the methods used. The total amount of proteins and phenols in the used membrane was only 0.45 mg/g, which was only 23% of the amount of total organics in the sample. Therefore a large portion of the organic foulants is still undefined. One of the main components of the cherry juice, the sugars, was not analysed, and its role in the fouling is therefore unknown.

The procedure used for the analysis of the fouled membrane proved to be good. However, some improvements can still be done. Over 70% of the organic foulants remained undefined, and it might be a good idea to analyse the sugars, which were one of the main components of the juice, and to do some microbiological analysis in order to check out the importance of biofouling. Moreover, the possible inorganic foulants could be analysed by EDAX.