Previous PageTable Of ContentsNext Page

New instruments to monitor and manage sprout damage

J.M.C. Dang1, M.L. Bason1, M.M. van Gasselt1, S.J. Neil1, and R.I. Booth1

1Newport Scientific Pty. Ltd., Warriewood, NSW, 2102, Australia.

Introduction

Sprout damage in wheat and barley results in increased α-amylase activity, which can cause detrimental effects on the quality of flour from wheat and the storage life of malting barley. Ideally, the α-amylase activity should be tested for each load of wheat or barley arriving at the silo. It is therefore important to devise a rapid and simple method for determining the α-amylase activity of the grains. The Hagberg Falling Number (FN) test and Newport Scientific Rapid Visco Analyser (RVA) Stirring Number (SN) test are two of the more common methods used to test α-amylase activity. Newport Scientific has recently developed a complete sprout damage testing solution, consisting of a prototype Clean Mill (CM) (Figure 1A) to reduce carry-over and cross contamination of samples, a new Robot Dispenser (RD) (Figure 1B) that automatically dispenses the corrected amount of water into a sample, and an RVA-StarchMaster (RVA-SM) (Figure 1C) that performs the SN test. This paper reports on the feasibility, accuracy and repeatability of the three instruments as a package for monitoring and managing sprout damage in wheat.

There are three main parts of the prototype CM through which the sample passes – the milling chamber, screen (and void), and tube. A hopper feeds whole grains to the milling chamber, which contains the impellor that grinds the grains. The grains are then passed through the screen to ensure uniformity in size. The ground sample then travels horizontally through a small straight tube before it is collected. The RD is designed to streamline the sample preparation process by weighing a sample of known moisture content, and automatically adds the corrected amount of water to the sample to keep the solids ratio constant. This allows the sample to be added by a simple volumetric scoop, simplifying sample preparation by automating the weighing and dispensing steps.

Figure 1. Prototype Clean Mill (A), Robot Dispenser (B), and RVA-StarchMaster (C).

Materials and Methods

Samples

Sound, whole, vitreous wheat grain was purchased from a local source, and subsampled to produce 250g samples for assessing preconditioning and sprouting effects as described below.

Preconditioning and Sprouting Wheat

The moisture content of the whole wheat was determined using AACC Method 44-15A (AACC, 2000), and the sample was designated “dry wheat”. A subsample (250 g) of whole wheat was preconditioned to raise the moisture by about 5%. Water was added slowly to the dry wheat using a pasteur pipette, with vigorous mixing after several drops to ensure thorough uptake of water by all grains. The wheat was then allowed to equilibrate for 24 h. The preconditioned wheat was designated “moist wheat”. To sprout wheat, excess water (~200 mL) was added to a subsample (250 g) of whole dry wheat. The wet sample was stored (sprouted) at room temperature for 72 h. The sprouted sample was then dried (30°C) for 72 h, and conditioned to 11.2% moisture content. This sample was designated “sprouted wheat”.

Grinding Samples on the prototype Clean Mill

Each whole wheat sample (100.0 g) was ground in duplicate on the prototype CM with Power Feeder, impellor, and 0.8 mm screen. To determine carryover, the weights of original (dry) and moist samples and their ground derivatives were accurately measured. Any ground material remaining in each of the three main sections of the mill was carefully brushed onto a tared container and weighed, such that losses from each section could be determined. Dry wheat was ground before moist wheat to avoid any carryover of moisture. To determine the effect of carryover on the SN, replicate samples of sprouted and original (dry) wheat were ground successively, without cleaning the mill between grinding, to allow residual material from milling the sprouted samples to contaminate the dry samples.

Rapid Visco Analysis

The moisture content of each ground sample was determined by AACC Method 44-15A (AACC, 2000), and reported as % as is. SN analyses were performed on the ground wheat samples using a Newport Scientific RVA-4 by AACC Method 22-08 (AACC, 2000). 4.00 ± 0.01 g of each ground sample (14% moisture basis) was added to 25.0 ± 0.1 g of distilled water (corrected for moisture content of sample) in a new canister. A paddle was used to disperse the sample, and the test started. The RVA curves of dry wheat that was ground before and after grinding of the sprouted wheat were compared. Repeatability was evaluated by one-way analysis of variance (ANOVA) of data, setting sample as the factor and using the error term for within-sample repeatability (Minitab Ver. 13).

Volume Allowance in RVA canister and Repeatability of RVA-SM and RD

The maximum and minimum allowances in total canister volume to obtain repeatable SN tests were determined using 4 samples of ground wheat with varying FNs. Using the RD, 4.00 g, 4.00 ± 0.10 g (± 2.5%) and ± 0.20 g (± 5%) of each sample was added to 25.0 ± 0.1 g of distilled water (adjusted for sample weight) in a new canister, and a stirrer was used to disperse the sample. The SN of each sample was determined on the RVA-SM using AACC Method 22-08 (AACC, 2000). Tests were performed in duplicate for repeatability. The effect of canister volume on SN results was evaluated by the General Linear Model (Minitab Ver. 13), setting sample as the factor (effect of sample removed) and the sample weight as a covariate. RVA-SM and RD repeatability were evaluated by one-way ANOVA.

Results and Discussion

Sample recovery and percentage losses from each section of the CM for dry and moist wheat are shown in Table 1. The CM retained only a small amount of ground sample for both dry and moist wheat. The milling chamber had low sample retention, with 0.02%, 0.03% and 0.08% for dry and moist wheat, respectively (Table 1). The tube had negligible sample retention. Of the three mill sections, the main contamination point was the area behind the screen, collecting 0.14%, 0.19% and 0.24% of ground dry and moist wheat, respectively. The moist wheat gave a slightly higher percentage of residual material.

RVA curves of sprouted wheat, and dry wheat ground before and after grinding the wheat, are shown in Figure 2, with Stirring Number values in Table 2. The high SN value of the dry wheat samples is typical of sound wheat (FN > 350). The low SN value of the sprouted samples indicates that they were highly sprouted. Nonetheless, milling sound dry wheat after milling the sprouted samples did not significantly reduce the SN values in these sound wheat samples (ANOVA F=3.04, P>0.05). These results indicate that the CM can be used without cleaning between samples when milling wheat for RVA SN evaluation.

Table 1. Milling losses on the Newport Scientific prototype Clean Mill.

Sample

MC (%)

Recovery (%)

Losses (%)

     

Chamber

Screen

Tube

Dry

11.63

98.2

0.021

0.139

0.003

Moist 1

14.22

96.9

0.029

0.188

0.007

Moist 2

15.26

96.3

0.079

0.244

0.003

Figure 2. RVA Stirring Number curves for dry sound wheat milled before (solid black lines) and after (dashed black lines) sprouted wheat (grey lines).

Table 2. Stirring Number results for dry sound wheat milled before and after sprouted samples, and for sprouted wheat.

Sample

Grinding Sequence

Rep

SN (RVU)

cP*

Dry Wheat 1

1

1

151

1816

   

2

152

1827

Dry Wheat 2

2

1

151

1816

   

2

148

1774

Sprouted Wheat

4

1

20

237

   

2

21

251

Dry Wheat 3 (contaminated)

5

1

148

1775

   

2

150

1805

Sprouted Wheat

6

1

19

223

   

2

20

236

Dry Wheat 4 (contaminated)

7

1

147

1764

   

2

149

1787

*cP = SN (RVU) x 12.

Table 3 shows SN results for wheat samples with varying FNs, tested on the RVA-SM using different sample weights (± 2% and ± 5%), and consequently different total canister volumes. For the sound wheat sample (Wheat1), using a sample weight of 2.5% below the recommended 4.00 g caused cavitation problems in SN tests. As the total slurry volume was reduced, air bubbles were more readily introduced into the slurry during the initial fast mix of the test. These bubbles remained behind the RVA paddle, unable to escape due to the thick nature of the swelled sample, resulting in unstable and non-repeatable results. These results were therefore not used in statistical analyses. Wheat samples 2, 3, and 4 were sufficiently less viscous such that any bubbles introduced into the slurry were able to escape during the test. Statistical analyses show that sample weight (canister volume) up to ± 5% did not significantly affect the SN results of the various wheat samples (ANOVA, F=1.24, p>0.05). With this larger sample weight tolerance of ± 0.2 g (± 5%) (previously recommended at ± 0.01 g), and the automatic adjustment and dispensing of water by the RD, sample preparation for SN analysis can simplified, allowing use of a volumetric scoop rather than a balance. However, it would be recommended that for all sample types (sound or sprouted), a minimum of 4.00 g sample is used to avoid cavitation problems.

Table 3. Effect of canister volume (sample and water weight) on Stirring Number results.

Sample Wt Class (g)

Rep

SN (RVU)

 

Wheat1

Wheat2

Wheat3

Wheat4

3.8

1

113.75*

80.50

27.42

7.33

3.8

2

108.00*

78.75

28.50

7.67

3.9

1

131.25

78.25

27.00

7.00

3.9

2

133.42

79.75

28.50

7.00

4.0

1

133.00

84.08

26.08

6.67

4.0

2

136.33

83.17

25.83

6.58

4.1

1

132.58

82.33

25.92

6.83

4.1

2

132.92

81.42

26.50

6.75

4.2

1

127.00

81.17

26.42

6.58

4.2

2

130.92

80.75

26.00

6.58

*Results treated as outliers during statistical analyses, due to cavitation problems.

The RVA-SM achieved good repeatability, with a RMS (root mean square) of 1.63 RVU and a coefficient of variation (CV) of 2.81%. From the measurements taken for the above SN tests, the RD achieved a mean accuracy of 0.027 g from the calculated target water weight, with a repeatability standard deviation of 0.045 g, or less than 0.05 mL, within recommended levels.

Conclusions

The prototype Clean Mill shows promise as a silo mill for preparation of samples for SN analyses, that does not require cleaning between samples. Elimination of the cleaning step would greatly reduce sample preparation time for SN analyses, and therefore increase sample throughput. A future modification to the prototype would include air assist behind the screen to further reduce carryover. The Robot Dispenser can accurately (accuracy of 0.027 g) and repeatably (repeatability STD 0.045 g) dispense the corrected amount of water into a sample, thereby simplifying sample preparation for SN analyses. The RVA-StarchMaster can perform SN tests on samples that have been ground and weighed by the CM and RD, respectively, giving repeatable results (CV of 2.81%) for wheat samples with a wide range of FNs. The Clean Mill, Robot Dispenser and RVA-StarchMaster show promise as a package for monitoring and managing sprout damage in wheat.

References

American Association of Cereal Chemists, Approved Methods, 10th edition. (2000) Approved Method 22-08.

American Association of Cereal Chemists, Approved Methods, 10th edition. (2000) Approved Method 44-15A.

Previous PageTop Of PageNext Page