FIELD OF THE INVENTION

The present invention relates to a method for mixing of at least two kinds of particles, particularly mixing of a small amount of one kind of particles with a larger amount of a different kind of particles.

BACKGROUND

Dry mixing of a small amount of one kind of particles into a larger amount of a different kind of particles is used, e.g., for mixing of enzyme particles into powder detergents or animal feed.

The current technology used to mix enzyme granules into powdered detergent products includes rotary drum mixers, which are able to mix small amounts of granular enzyme products efficiently without causing damage to the enzyme granules. Drum mixers are, however, large and costly process equipment. In addition is takes up substantial amount of space in the detergent manufacturing plants. This makes the introduction of enzymes into new plants costly and difficult. Mixing of liquids with liquids or mixing of gases are on the contrary done very easily and efficiently using so-called static mixers, which are very compact and inexpensive equipment and used for many years throughout the process industry. Application of static mixers for mixing powders are, however, very scarce. Only few papers have been published on attempt to use static mixers for mixing dry or essentially dry powder streams. Powders are inherently more difficult to handle than liquids. The multi-phase nature of powder flows are still not very well understood despite the advancement of sophisticated modelling and simulation tools becoming available; such as CFD (Computational Fluid Dynamics) and DEM (Discrete Element Method). No single theory is, however, able to accurately describe powder flows. These difficulties have led the industry to staying with current mixing technology such as rotating drum and paddle mixers.

Generally the same technology is used to produce blends of two or more kinds of enzyme particles, e.g. enzyme granules with different kinds of detergent enzymes.

Static mixers have the drawbacks of being difficult to apply for powder mixing due to the lack of understanding of powder flows and mixing. In addition the mechanical constructions contain by nature sharp mixing elements, which potentially could damage sensitive particles. It is also known that the mixing quality is strongly dependent on the flow patterns in the mixer, the relative densities and particle size distributions of the powders to be mixed. SUMMARY OF THE INVENTION

The inventors have found that continuous mixing in a static mixer can be used to add one kind of particles (such as an enzyme granular product) in a small amount to a larger amount of a different kind of particles (such as a powder stream of detergent powder), even if the powder characteristics are substantially different, with substantially no damage to the enzyme particles without and with a high degree of homogeneity.

Accordingly, the invention provides a method for mixing of at least two kinds of particles, wherein a first and a second stream of particles are mixed in a static mixer, and wherein the first and the second stream of particles have a weight ratio below 1 :20.

The mixing quality (or mixing homogeneity) of the two kinds of particles is a key feature of the invention. Thus, in the resulting mixture of the first and second stream of particles, the coefficient of variation of the amount of particles from the first stream, preferably enzyme particles, is less than 50%, preferably less than 40%, more preferably less than 30%. The coefficient of variation may be measured using at least 9 samples and a sample size of approximately 5 g, and preferably 0.3 wt.% particles from the first stream. If an analytical method is used to determine the amount of particles from the first stream, the coefficient of variation of the analytical method must be subtracted from the measured coefficient of variation.

Another key feature of the invention, when the first stream of particles is enzyme particles, is that the level of free enzyme dust in the resulting mixture is surprisingly low, as shown in the Examples. Thus, the resulting mixture of the enzyme particles and the second stream of particles may have an amount of free enzyme dust below 20 parts per billion, using the standard fluidization assay (J. Liu et al., Building and Environment, 44 (2009), 2327-2334).

The invention also provides a method for mixing of at least two kinds of enzyme particles, wherein a first and a second stream of enzyme particles are mixed in a static mixer.

DETAILED DESCRIPTION

First particle stream

The first particle stream typically comprises one or more kinds of enzyme particles or other active or functional particles, such as bleach activators, polymers, colorants, perfume particles etc. The enzyme particles typically have particle sizes in the range 100-2000 μηι, preferably an average particle size in the range 300-1200 μηι. They generally consist of a core comprising the enzyme surrounded by a coating. The core may additionally comprise finely divided cellulose fibers. The coating may comprise a wax such as polyethylene glycol or a salt such as sodium sulfate. Some formulations of enzyme particles are described in US4106991 , W09212645, WO0001793, WO0104279, WO06034710, WO201 1 134809, WO200024877, WO 2003000625 and WO2007044968. Second particle stream

As examples, the second powder stream may consist of powder detergent (granular detergent), animal feed or flour premix for baking where it is of interest to add minor ingredients in very small dosages. The particles should be free flowing as determined by an angle of repose being less than 45°, particularly less than 40° or less than 35°.

Thus, the invention may be used to add minor ingredients such as enzyme particles to powder detergent (granular detergent), Relevant enzymes include one or more of the following: a protease, an amylase, a carbohydrase, a lipase, a cellulase, an oxidoreductase, a mannanase and a pectate lyase. Examples are described in US4106991 , W09212645, WO0001793, WO0104279, WO201 1 134809, WO200024877 or WO 2003000625.

Another example is the addition of feed enzymes to animal feed. Relevant enzymes include one or more of the following: a protease, an amylase, a carbohydrase, a lipase, a cellulase, and a phosphatase such as phytase. Examples are described in W09212645, WO 2001/004279, WO06034710, WO2007044968, WO9623062.

The invention may also be used to add enzymes to a flour premix for baking. Relevant enzymes include alpha-amylase. maltogenic amylase, lipase, phospholipase, xylanase.

Examples are described in W09104669, WO9826057.

Mixing of enzyme particles

A static mixer can also be used for mixing of two or more kinds of enzyme particles.

Examples of relevant enzymes are those listed above.

Static mixer

The static mixer is a device which can provide continuous mixing of streams of solid particles using no moving components. It may be a helical mixer consisting of mixer elements housed in a cylindrical tube, where the static mixer elements consist of a series of baffles. The static mixer should have a diameter which is at least 10 times of the particle size.

EXAMPLES

Example 1 -3: In-line mixing of enzyme into detergent powder

Streams of two types of particles were used, as follows:

(a) Detergent: commercial European powder detergent without enzyme.

(b) Enzyme: Commercial protease granules (Savinase™, product of Novozymes A/S). The average particle size is in the range 300-1200 microns. The equipment for dosing and mixing was as follows:

The detergent powder and the enzyme granules were each dosed by a vibrating feeder giving essential constant volumetric flow rates. A flow rate of 20 kg/min (+/- 1 kg/min) was used for dosing the detergent powder and the enzyme feeder set-point was set to dose 16, 31 , and 62 g enzyme/min., corresponding to with dosages of 0.15%, 0.3 and 0.6% of enzyme by weight, respectively.

Measurements were done as follows:

Full flow samples were collected over 1 minute corresponding to about 20 kg. The homogeneity of the mixing was measured by scooping out nine samples (2a, 2b, 2c, 4a, 4b, 4c, 6a, 6b, 6c) of -5 g scoop size randomly from each test. It was secured that no additional mixing took place during the sampling procedure.

The results were as shown in the following Tables.

Table 1 .

Example-2-2b 0.3 5.19 0.1 100

Example-2-2c 0.3 5.18 0.0988

Example-2-4a 0.3 5.22 0.0726

Example-2-4b 0.3 5.24 0.1080

Example-2-4c 0.3 5.24 0.1310

Example-2-6a 0.3 5.16 0.0832

Example-2-6b 0.3 5.18 0.0801

Example-2-6c 0.3 5.21 0.0681

Table 3.

Table 4. Mixing homogeneity as function of dose.

Enzyme granules Mean Std. Dev. Cv Cv (scoop size: 5 g)

measured minimum

0.15 wt.% 0.049 0.014 28% 24.7%

0.3 wt.% 0.092 0.021 23% 17.5%

0.6 wt.% 0.273 0.077 28% 12.4% The mixing quality is determined by doing activity analysis of a number of scoops of product taken randomly. The scoop size is selected according to the application.

The best mixing quality that is possible to achieve corresponds to a statistical random distribution of the enzyme particles among the particles of the major powder stream(s). From the particle size distribution it is possible to calculate the distribution of the number of particles which is in a completely randomly mixed product for a given dosage and a given sample size (also called scoop size). When the distribution is known the standard deviation of the total amount of enzyme present in a scoop may be calculated. The average value will be equal to the dosage, so the Cv (coefficient of variation) can be calculated as the standard deviation divided by the average value. In Table 4, this Cv (Cv min) has been calculated for the three different dosages and a scoop size of 5 g. It is seen that the measured Cv, which includes the fixed value of the analytical Cv of the enzyme assay, gets close to the best possible Cv - especially for the smaller enzyme dosages.

The amount of free enzyme dust that is present in a sample is measured using a standard fluidization assay (J. Liu et al., Building and Environment, 44 (2009), 2327-2334). Table 5 shows the obtained results for the three tested dose levels: 0.15, 0.3 and 0.6 wt.% enzyme. The results are plotted in units of PPB (parts per billion or ng (10 ^"9 g) active enzyme/g product. The blank-level of the method is -2 ng/g, and typical commercial laundry powders have active enzyme dust levels in the range 5 - 20 ng/g. Thus, the results show an extremely low level of active dust in all samples.

Table 5.

Example-2-4b 0.3 1490 7.26

Example-2-6a 0.3 966 2.47

Example-2-6b 0.3 1000 2.48

Example-3-2a 0.15 861 1 .55

Example-3-2b 0.15 668 3.28

Example-3-4a 0.15 1200 1 .93

Example-3-4b 0.15 990 2.35

Example-3-6a 0.15 1020 0.98

Example-3-6b 0.15 1 180 2.26

The results demonstrate efficient mixing to a high degree of homogeneity, with a coefficient of variation close to the theoretical value for the sample size and particle size in question. The results also demonstrate an extremely low dust formation, well below the commonly accepted limit of 60 ng/m ³ .