NIR Technology Australia        

 

 

                                                                                    Suite 103, 56 Kitchener Pde

                                                                                                                                                Bankstown, NSW, 2200

                                                                                                                                                Ph: 02 9708 5068

                                                                                                                                                Fax: 02 9708 5537

 

 

 

Application note 32: Cropscan 2000B, Preliminary Evaluation of Protein Analysis in Sorghum.

 

Sample

Predicted Y

Measured Y

Difference

 

 

 

 

 

RT007

6.56

6.59

0.03

 

 

 

 

 

30208252

7.28

7.17

-0.11

 

 

 

 

 

10

8.12

7.88

-0.24

 

 

 

 

 

30208021

8.49

8.44

-0.05

 

 

 

 

 

2

8.71

8.78

0.07

 

 

 

 

 

H2399

9.04

9.39

0.35

 

 

 

 

 

4

9.99

9.94

-0.05

 

 

 

 

 

30208028

10.35

10.67

0.32

 

 

 

 

 

H2402

11.01

11.08

0.07

 

 

 

 

 

110829

11.68

11.57

-0.11

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Bias

0.03

 

 

 

 

 

 

 

Std Dev

0.19

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Procedure:      Approximately 50 sorghum samples scanned on Cropscan 2000B in an 18mm pathlength cell, 5 replicates per sample.

Sorghum protein determined by Bunge Feedstock laboratories by Kjeldahl analysis.

Preliminary calibration developed on 10 samples covering a wide range of protein content. Partial least squares regression was the method used and the calibration statistics are shown in the results section for the average of the 5 replicates.

 

 

 

 

 

Results:

Conclusion:     The above results show that the Cropscan 2000B whole grain analyser can be calibrated to measure the protein content of sorghum samples, which come from a wide variety of types and seasons. Slight adjustments to the pathlength and more data should lead to a robust calibration. Moisture analysis is already known to be successful in sorghum with prediction statistics of about 0.2%.

 

 

 

 

Brad Swarbrick

Applications Chemist

 

29/11/2001