Analyses of HQCF quality at NRI in 2012 Aurelie Bechoff & Louise - PowerPoint PPT Presentation

Analyses of HQCF quality at NRI in 2012 Aurelie Bechoff & Louise Abayomi

Samples analysed Location Variety(s) of Production Collection Company Drying method Analysis (Country) cassava used date date Dokuduade Ghana Caltech Ventures flash 15-Feb 23-Feb 01-Mar (improved) local variety (red Cassacosa sun (3 days) 22-Feb 23-Feb 01-Mar skin) hot-air tray dried for St Baasah mix of varieties 22-Feb 23-Feb 01-Mar 2h & sun dried for 6h Nigeria Thai Farm unknown flash 14-Feb 21-Feb 29-Feb Eagle Baba unknown flash 17-Feb 17-Feb 29-Feb MicMakin unknown flash 15-Feb 16-Feb 29-Feb Peak Products unknown flash 15-Feb 22-Feb 29-Feb Tanzania Daldo Nat Wara unknown sun Jan 31-Jan 23-Feb Kejo Ent Masasi unknown sun Jan 02-Feb 23-Feb Ungawo Muhogo unknown sun Jan 31-Jan 23-Feb Uganda PATA NASE 1 sun 23-Feb 24-Feb 05-Mar SOSPA NASE 3 sun 24-Feb 25-Feb 05-Mar Malawi CMRTE unknown sun 07-Feb 15-Feb 19-Mar Tiamike unknown sun 23-Nov-11 17-Feb 19-Mar

HQCF is: • White in colour • Bland in taste, not sour • Low in cyanide (<10mg/Kg) • Has particle size <0.25mm • Odorless • Free of foreign matter • Free of mould, with low microbial count • Has a moisture content of 10-12% • Not fermented (pH >5.5) The Process high Quality Cassava Flour (HQCF): STEP 1. Always use good quality cassava roots; this is to ensure optimum recovery of HQCF and white colour STEP 2. Within 12 hours of harvest peel roots removing any rots; this is to minimize waste-roots will start to deteriorate after 12 hours of harvest STEP 3. Wash peeled roots in clean water; this is to achieve low microbiological levels in flour STEP 4. keep peeled washed roots in clean water for short period (1 hour) until grating; this is to avoid discoloration and ensure white colour STEP 5. Produce a wet mash by grating (never use a chipper!); this is to facilitate water removal and also minimize cyanide levels in flour STEP 6. Remove as much water from grated mash as possible using a hydraulic press; this is to speed up the drying process STEP 7. Dry pressed cassava cake as quickly as possible (6 hours) and completely this is to avoid fermentation, bad odor, sour taste and mould STEP 8. Mill dried cassava grits to 0.25mm this is to produce a fine flour STEP 9. Sieve to produce fine free flowing flour this is to produce a fine free flowing flour and remove foreign matter STEP 10. Bag in polypropylene bags with liner and keep in hygienic dry aerated store this is to prolong the quality THE WHOLE PROCESS MUST BE COMPLETED WITHIN 24 HOURS

Specifications for HQCF analyses  White in colour  Bland in taste, not sour • Colour  Low in cyanide (<10 mg/kg)  90% has particle size <250 um • pH  Odorless  Free of foreign matter • Particle size  Free of mould with low microbial count • Moisture  Has a moisture content of 10 ‐ 12%  Not fermented (pH > 5.5) content • RVA

Orange = quality needs improvement Green = quality ok

Colour Location Company Colour (Country) Ghana Caltech Ventures yellowish Cassacosa yellowish St Baasah yellowish Nigeria Thai Farm white Eagle Baba white MicMakin white Peak Products white Tanzania Daldo Nat Wara yellow Kejo Ent Masasi not pure white Ungawo Muhogo yellow Uganda PATA yellowish SOSPA white Malawi CMRTE white Tiamike white

pH Location Company pH value (Country) Ghana Caltech Ventures 5.7 Cassacosa 7.0 St Baasah 6.2 Nigeria Thai Farm 6.0 Eagle Baba 6.2 MicMakin 4.9 Peak Products 4.4 Tanzania Daldo Nat Wara 5.7 Kejo Ent Masasi 5.8 Ungawo Muhogo 6.0 Uganda PATA 6.3 SOSPA 6.5 Malawi CMRTE 6.2 Tiamike 5.7 pH value was measured in triplicate

Moisture content Location Company Moisture content % (Country) Ghana Caltech Ventures 12.8% Cassacosa 14.7% St Baasah 8.2% Nigeria Thai Farm 10.1% Eagle Baba 8.5% MicMakin 7.0% Peak Products 13.8% Tanzania Daldo Nat Wara 15.1% Kejo Ent Masasi 12.4% Ungawo Muhogo 14.6% Uganda PATA 8.9% SOSPA 6.9% Malawi CMRTE 15.1% Tiamike 14.3% Moisture content was measured in triplicate

Particle size Location Particle size (% Company (Country) <0.25um) Ghana Caltech Ventures 78.9% Cassacosa 77.0% St Baasah 91.0% Nigeria Thai Farm 95.4% Eagle Baba 96.3% MicMakin 95.5% Peak Products 93.4% Tanzania Daldo Nat Wara 54.1% Kejo Ent Masasi 74.7% Ungawo Muhogo 77.8% Uganda PATA 93.6% SOSPA 98.9% Malawi CMRTE 96.7% Tiamike 96.7% Particle size (by sieving) was measured in duplicate

RVA: rapid visco ‐ analyser Starch properties For samples destined to adhesive paperboad or plywood industries.

RVA profile

Example: profile of SOSPA vs PATA ‐ Uganda 3500 SOSPA 3000 PATA 2500 2000 Viscosity (cP) 1500 1000 500 0 0 500 1000 1500 ‐ 500

Example: profile of Thai Farm, Peak Products & Caltech 3500 Peak Products 3000 Thai Farm 2500 Caltech 2000 Viscosity (cP) 1500 1000 500 0 0 500 1000 1500 ‐ 500

RVA Location Peak Final Pasting Company Trough Breakdown viscosity Setback (Country) viscosity cP temp °C Ghana Caltech Ventures 1769.0 123.5 1645.5 285.0 161.5 67.9 Cassacosa 1780.5 544.5 1236.0 880.5 336.0 67.8 St Baasah 1956.5 810.5 1146.0 1246.5 436.0 69.1 Nigeria Thai Farm 3082.5 1127.5 1955.0 2136.0 1008.5 65.9 Eagle Baba 2697.0 862.0 1835.0 1498.0 636.0 69.2 MicMakin 1821.0 566.5 1254.5 912.5 346.0 68.0 Peak Products 1977.0 589.5 1387.5 918.5 329.0 70.5 Uganda PATA 1824.5 616.5 1208.0 1004.0 387.5 66.8 SOSPA 3059.0 1061.0 1998.0 1630.5 569.5 64.8 Malawi CMRTE 2326.5 1344.5 982.0 2090.5 746.0 70.8 Tiamike 2669.0 1259.0 1410.0 1945.5 686.5 67.1 RVA was done in duplicate

Conclusion  Not all the HQCF samples meet the specificities – still effort to make on the quality  Some sun ‐ dried sample (e.g. SOSPA) very good quality – therefore sun ‐ drying not a obstacle to quality (although the size of production is limited)

Way forward  Put into place a quality control system in the different countries ‐ all these analyses easy to carry out but RVA

Where are we?  Across C: AVA countries, only 20% (3 of 15) of samples met the full HQCF specification  2 (of 4 samples sent) from Nigeria, and 1 (of 2 samples sent) from Uganda  57% met particle size spec.  86% met pH spec.  58% met moisture content spec.  67% met foreign matter spec.  86% met odour spec.  57% met fibre content spec.  87% met colour spec.

What have we done?  Local training in TQM/Quality Management (Nigeria, Ghana, Tanzania)  Quality management workshop/demonstration with measurement tools ‐ particularly particle size and pH, traceability and documentation (Uganda, Malawi)  Tanzania ‐ No field audit

What have we done?  Uganda ‐ Field audit undertaken with follow up  Malawi ‐ Field audit undertaken with follow up  Ghana ‐ No field audit  Nigeria ‐ Field audit undertaken, no follow up  HQCF samples tested in ‐ country/ at NRI (few)

Why are we here?  Lack of tools for assessing quality  Incorrect screens within mill  Poor peeling and/or opportunity for product contamination  Non ‐ adherence to recommended processing times ‐ leading to fermentation

Why are we here?  Overmature roots  Insufficient pressing/drying  Lack of focus on importance of quality (Team)  Most issues appear in countries with no audit (i.e. Ghana and Tanzania)

Where do we want to be?  Positive feedback from end ‐ users  Fit for purpose ‐ a quality product fit as a partial substitute for wheat flour/substitute for corn starch  Consistent quality  <5% total defects is the norm  Achieve the same quality standard across processors and ALL countries

How do we get there  Not just about training ‐ but attitude!  May need to hand ‐ hold (processors) ‐ initially  Need for simple tools (processors and end ‐ users) for objective measurement of main parameters (particle size and pH)

How do we get there?  Is there sufficient time to raise standards?  Next 6 months critical  Obtain regular feedback from end ‐ users by BDA  Scheduled analyses by NRI  Scheduled analyses by CM’s  In ‐ house (processors) quality control  NRI backstopping  Other

How do we get there  Need to ensure correct screens available  Increased technical backstopping  Training on equipment  Monitor output volumes  CM’s need to spot check (processors) ‐ until confident ‐ all CMs/BDA should have 250  m sieve and pH measurement tool

How do we get there  Bring main processors and end ‐ users together to reaffirm what HQCF is ‐ visibly objectively confirming particle size and pH ‐ only once you can achieve it!  Quality control during peeling, improved handling practices/processing environment to eliminate presence of ‘black spots’  Eliminate cross contamination during milling

How much will it cost?  250  m sieve (UK) = £150  pH strips (UK) = £15  Cost of training/backstopping by NRI = 1 flight, 5 days = £3, 500  HQCF sample analysis (NRI) = £5000/country/yr  Screens for mill = ?  In ‐ country training costs = (based on Malawi, 20 people x 2days) = £8, 000  Processors = daily labour rate