Canadian Grain Commission
Symbol of the Government of Canada

The implications of frequently encountered grading factors on the processing quality of durum wheat

4. Factors affecting processing

Test weight

Test weight is a widely recognized primary specification in wheat grading because it is related to the degree of soundness of wheat. Test weight is often used as an index of milling potential, but there is no consensus on its true value as a milling yield predictor (Hook et al. 1984). Different wheat classes and different varieties within a wheat class exhibit different relationships between test weight and milling yield. Test weight is affected by moisture content, weathering, kernel size and density, and packing factors which have little direct relationship to milling potential.

Test weight is a useful index for durum wheat milling potential. Dexter et al. (1987) found a strong relationship between CWAD semolina yield and test weight for durum wheat for two consecutive crop years (Figure 1A). Increasing semolina ash content with decreasing test weight caused a dramatic drop in semolina milling score (semolina yield at constant ash content) as test weight declined (Figure 1B). Watson et al. (1977) also concluded that test weight was an effective indicator of milling potential for American durum wheat.

Relationship of test weight of CWAD to semolina yield and to semolina yield at constant ash content

Figure 1. Relationship of test weight of CWAD to semolina yield (A) and to semolina yield at constant ash content (milling score, B). Source: Dexter et al. (1987).

Protein content

Protein content is the most important quality characteristic influencing pasta cooking quality (Matsuo et al. 1982a; Autran et al. 1986; D'Egidio et al. 1990). The relationship is complex and is influenced by other factors, notably protein content and processing conditions, but generally, as protein content increases, pasta becomes firmer and less sticky. Pasta made from high protein semolina is physically strong and elastic. High protein cooked pasta is firm, non-sticky and resilient, and retains its texture if overcooked. Pasta produced from low protein semolina is deficient in some or all of these characteristics.

The relationship of protein content to pasta cooking quality is evident from CGC CWAD harvest survey data over the past 10 years (Table 3). High rainfall in western Canada over the past five years has low protein content, due to high wheat yields. The previous five years were drier, and protein content was higher. Pasta cooking scores were clearly superior during the higher protein period.

Durum wheat is often traded at guaranteed minimum protein content levels. Countries such as Japan, the United States and Spain have legislation and/or labeling laws that stipulate minimum protein content levels in the finished product. In Canada low protein content in recent harvests has necessitated segregating the top three grades of CWAD by protein content to meet minimum specifications of discriminating customers. However, unlike protein segregation of wheat classes such as hard red spring wheat, protein segregation of durum wheat is not the complete solution to demands from the marketplace. Medium to low protein hard red spring wheat can be used for high quality products (hearth bread, Asian noodles etc.). In contrast, low protein durum wheat remaining after segregation of high protein material has low market acceptance and lower value because it has limited application outside the pasta industry.

Table 3 – No. 1 CWAD wheat protein content and pasta cooking score (cooking quality parameter, CQP)
Harvest Year Protein % CQP units Harvest Year Protein % CQP units
1997 12.7 33 1992 13.0 37
1996 12.4 33 1991 12.6 46
1995 12.1 38 1990 14.1 56
1994 12.5 36 1989 15.0 57
1993 11.9 28 1988 15.5 54
5 year mean 12.3 34 5 year mean 14.0 50

Source: Canadian Grain Commission Harvest Survey Data.

Abbreviations: CQP = cooking quality parameter, a score combining cooked pasta firmness and resilience.

Yellow Berry (nonvitreous, starchy or mealy kernels)

Yellow berry (mealy kernels, starchy kernels) is a physical property of wheat that is a primary durum wheat marketing factor. Starchy kernels occur when insufficient nitrates are available during grain development. Starchy kernels are important because they are lower in protein content and softer that vitreous kernels (Dexter and Matsuo 1981; Dexter et al. 1988; Dexter et al. 1989a).

Starchy zones in wheat kernels are opaque and appear chalky when cut (Figure 2). This is due to air pockets in starchy zones. Vitreous zones have a compacted continuous structure, with starch granules tightly bound within a protein matrix. Starchy zones are less compacted, implying less protein than in vitreous zones. However, protein content and protein composition of vitreous and starchy zones in piebald kernels are similar (Dexter et al. 1989a). Starchy zones may occur in piebald kernels because there is insufficient protein to create a sufficiently strong matrix to fully contract all regions during final stages of maturity as kernels desiccate.

The structure of piebald (partially starchy) durum wheat kernels

Figure 2. The structure of piebald (partially starchy) durum wheat kernels. Top, external view showing chalky appearance of starchy zone. Bottom, scanning electron micrograph showing starchy-vitreous interface. Source: Dexter et al. (1989a).

As discussed earlier, durum wheat often is marketed with a minimum protein content guarantee. When protein content is specified hard vitreous kernel content is less important, but the softer texture of starchy kernels is still a milling factor. The relationship between starchy kernels and durum wheat milling performance is complex, but generally starchy kernels yield less coarse semolina and more flour, reducing durum wheat milling potential in markets where coarse semolina is preferred.

Bolling and Zwingelberg (1972) found that durum wheat semolina yield was more related to wheat origin than to wheat vitreousness. Internationally recognized procedures (ISO and ICC) define fully vitreous kernels as “those that do not disclose the least trace of farinacious endosperms”, a definition adopted by the CGC. Matveef (1963) proposed that fully starchy kernels should be given more emphasis that piebald kernels when estimating durum wheat vitreousness. Piebald kernels are lower in protein content that vitreous kernels, but they are almost as hard as vitreous kernels (Table 4). Fully starchy kernels are significantly softer, and cause lower yield of coarse semolina.

The impact of starchy kernels on semolina yield varies from miller to miller, depending on semolina granulation targets. Over 25 years ago Menger (1971) questioned the importance of starchiness on durum wheat milling, given a tend to finer semolina granulation. Since then the trend to finer granulation has become more apparent, as major pasta equipment manufacturers recommend finer granulation for modern high capacity presses.

Table 4 – Protein content and hardness (particle size index) of vitreous, piebald and starchy kernels from three durum wheat cultivars
Sample Protein % PSI %
Vitreous 11.8 28.6
Piebald 10.0 33.7
Fully starchy 9.7 46.2
Vitreous 10.7 36.6
Piebald 8.8 37.8
Fully starchy 7.9 48.5
Vitreous 11.8 28.6
Piebald 10.0 33.7
Fully starchy 9.7 46.2

Source: Coulter: Dexter et al. (1988); Wascana and Wakooma: Dexter et al. (1989b).

Orange wheat blossom midge

In our previous article (Dexter and Edwards 1998) we summarized the very serious quality problems that occur when the orange wheat blossom midge (Sitodiplosis mosellana Géhin) attacks common wheat. Orange wheat blossom midge eggs are deposited on florets during heading and flowering, and larvae feed on developing grain. Severe midge infestation has devastating effects on crop yield unless insecticide treatments are applied soon after infestation is apparent. Grain from midge-damaged common wheat exhibits high protein content, poor milling quality and weak gluten.

In 1995 and 1996 midge damage was a major grading factor for CWAD and American durum wheat. Gluten quality is an important secondary quality factor for pasta cooking quality (Feillet and Dexter 1996). However, the high protein content associated with midge damage could offset the adverse effect of midge damage on gluten strength.

The CGC conducted an extensive survey following the 1995 harvest to evaluate the importance of midge-damage on CWAD processing performance (Dexter and Marchylo unpublished data). Results confirmed that the protein content of midge-damaged durum wheat tended to be high. There was little evidence that gluten quality, measured by the SDS-sedimentation test and wet gluten yield, was damaged (Table 5). As a result pasta texture was not affected.

Midge damage had a serious effect on durum wheat semolina milling performance (Table 5). As midge damage increased, semolina refinement (ash content, color and speck count) declined and pasta became less bright and undesirably brown (long dominant wavelength). Yellow pigment in semolina and yellowness of pasta (purity) were not affected. Severely midge damaged kernels, which were rotted and blackened, presumably from attack by fungi, were particularly detrimental to semolina refinement and spaghetti color.

Table 5 – Effect of midge damage and severe midge damage on the milling and pasta-making quality of CWAD wheat from the 1995 harvest
Property Sample 1 Sample 2 Sample 3
Midge damage, % 3.7 6.9 6.7
Severe midge damage, % 0.4 1.8 8.6
Test weight, kg/hL 81.5 80.7 80.5
Protein content, % 14.2 14.3 14.5
SDS-sedimentation, mL 32 34 31
Semolina yield, % 66.1 66.5 67.3
Protein content, % 12.9 13.3 13.2
Wet gluten, % 32.2 32.6 33.1
Ash content, % 0.65 0.69 0.71
AGTRON color, % 68 62 58
Yellow pigment, ppm 7.5 7.5 7.6
Specks per 50 cm2 58 70 126
     Brightness, % 47.6 45.5 43.9
     Purity, % 52.4 51.5 51.3
     Dominant wavelength, nm 577.2 577.7 578.2
Cooking score, units 70 69 71

Source: Dexter and Marchylo, internal Canadian Grain Commission report. Analytical values on 14% mb.

Frost damage and immaturity

The short growing season in western Canada and the northern United States makes wheat vulnerable to frost damage and immaturity. Effects of frost damage on common wheat has received considerable attention (Dexter and Edwards 1998 and references therein). Severe frost damage is one of the most serious quality defects associated with common wheat because of adverse effects on wheat milling performance and baking quality.

Dexter and Matsuo (1981) examined the effects of immaturity and moderate frost damage on the milling and pasta-making quality of CWAD wheat using hand-picked samples from the 1979 harvest (Table 6). They found that the main effects were loss of semolina refinement (high ash content and more specks) and duller and browner spaghetti (longer dominant wavelength).

Table 6 – Effect of moderate frost damage on milling and spaghetti properties of CWAD wheat
Property No. 1 CWAD No. 3 CWAD
Light frost, % 0 10
Test weight, kg/hL 84.1 81.3
Protein content, % 14.7 15.0
Semolina yield, % 68.6 66.7
Protein content, % 13.8 14.0
Ash content, % 0.68 0.73
Specks per 50 cm2 13 20
     Brightness, % 45.2 43.4
     Purity, % 59.4 58.2
     Dominant wavelength, nm 577.9 578.1
Cooking score units 14 17

Source: Dexter and Matsuo (1981). A sample of 1979 crop CWAD wheat was hand-picked to give a frost-free sample (No. 1 CWAD) and a sample enriched in frost damaged kernels (No. 3 CWAD).

In 1992 frost damage and immaturity were the predominant grading factors associated with the Canadian durum wheat crop. That permitted Dexter et al. (1994) to prepare a series of samples grading from No. 1 CWAD to feed quality (No. 5 CWAD) solely due to frost damage and immaturity. They verified that increasing immaturity and moderate frost damage caused a gradual decline in semolina refinement and a concomitant deterioration in spaghetti color.

Gluten properties were affected by severe frost damage. CWAD downgraded to No. 3 on account of frost damage exhibited a Mixograph curve comparable to No. 1 CWAD because severe frost damage is not allowed in No. 3 CWAD (Figure 3). CWAD downgraded to No. 4 CWAD, which contains some severe frost damage, gave a significantly weaker curve, and No. 5 CWAD gave an abnormal curve. Due to the inclusion of some severe frost damage in No. 4 CWAD there was a pronounced deterioration in semolina milling yield, semolina refinement and pasta color, but cooking quality was not greatly affected (results not shown).

Semolina mixograph mixing curves.

Figure 3. Semolina mixograph mixing curves for No. 1 CWAD and samples downgraded to No. 3, No. 4 and No. 5 CWAD solely on the basis of frost damage and immaturity. Source Dexter et al. (1994).

Sprout damage, weathering and mildew

Pre-harvest sprouting due to damp harvest conditions has serious adverse effects on bread quality (Dexter and Edwards 1998 and references therein) The starch degrading enzyme alpha-amylase is present in very high levels in sprouted wheat. Alpha-amylase degrades starch during mixing and fermentation reducing the water holding capacity of starch which causes lower baking absorption and dough that is sticky and hard to handle.

The effect of sprout damage on durum wheat processing quality is less obvious. There is general agreement that sprout damage alone has little if any effect on semolina milling performance (Donnelly 1980; Matsuo et al. 1982b; Dexter et al. 1990). Debbouz et al. (1995) found that minor bleaching (weathering) caused by rain at harvest, involving only discoloration of the seed coat, did not affect semolina properties or pasta color.

Sprout damage is often associated with mildew damage, caused by a fungus (Cladosporium) which poses no toxicological threat, but can cause semolina milling problems. Light mildew damage is characterized by grey tufts of spores at distal ends of damaged kernels. Severely mildewed kernels become blackened and rotten. Unless they are removed during cleaning moderately to severely mildew damaged kernels in durum wheat cause dark specks in semolina (Dexter and Matsuo 1982).

Deleterious effects of sprout damage on bread making quality are exacerbated by degradation of starch by alpha-amylase during fermentation. The lower water content of pasta dough and rapid loss of moisture during drying provides alpha-amylase with little opportunity to degrade starch during processing (Dexter et al. 1990). During cooking, alpha-amylase is heat denatured rapidly by penetration of cooking water (Kruger and Matsuo 1982).

Canadian (Matsuo et al. 1982), French (Combe et al. 1988) and American (Dick et al. 1974) studies have concluded that sprout damage has little effect on pasta texture. As durum wheat Falling Number (FN) declines to about 150 seconds, there is no effect on pasta cooking quality (Table 7). At lower FN, which corresponds to sprout damage far in excess of levels encountered in durum wheat marketed for high quality pasta, there are slight influences on pasta cooking quality and amount of solids lost to cooking water. Production problems such as uneven extrusion, strand stretching and irregularities in drying (‘checking’ or cracking of strands during storage) have been attributed to sprout damage, but only when damage is very severe (Donnelly 1980; Combe et al. 1988).

Table 7 – Effect of sprout damage on durum wheat Falling Number, semolina a-amylase activity and pasta cooking quality
CWAD grade Sprout damage, % Wheat Falling Number, econds Semolina a-amylase activity, units Spaghetti cooking loss, % Spaghetti firmness, nm/sec
1 0 360 34 6.2 37
2 1.3 250 144 5.9 34
2 2.2 216 193 5.6 36
3 3.5 196 350 6.5 38
3 7.0 179 402 6.3 38
4 10.4 101 876 6.9 46
4 12.5 80 1363 8.0 50

Source: Matsuo et al. (1982). Firmness measured as rate of shear, lower values indicating greater firmness.

Heat damage

Wheat growing areas of North America usually are favorable for drying grain naturally in the field. However, occasionally wet harvests can result in damp grain. Improper storage of damp grain, or artificial drying at too high a temperature can cause heat damage. In extreme cases kernels turn black and emit a charred odor (binburnt kernels). Binburnt kernels are a serious quality defect in durum wheat even at low levels because of highly visible dark specks in semolina.

Binburnt kernels are readily detectable visually. However, damage to gluten functionality by artificial drying may occur without visual evidence. Heat damage has little effect on semolina refinement or pasta appearance (Dexter et al. (1989b). However, heat damage to gluten proteins influences physical dough properties, and when severe, can result in unsatisfactory pasta cooking quality (Figure 4).

Effect of severe heat damage and moderate heat damage on durum wheat spaghetti cooking properties

Figure 4. Effect of severe heat damage (A,B,C) and moderate heat damage (D,E,F) on durum wheat spaghetti cooking properties. Source: Dexter et al. (1989b).

Durum wheat heat damage can be detected from mixing curves. Moderately heat damaged durum wheat exhibits pronounced delay in reaching peak dough consistency, and peak consistency is reduced. A rapid micro-milling and mixing procedure adapted from the method of Preston et al. (1989) is effective in detecting heat damage in durum wheat (Dexter et al. 1989b).

Preston and Symons (1993) developed a rapid simple heat damage test that estimates the extent of protein fibril formation when a ground sample is wetted. When viewed by bright field microscopy the extent of fibril formation clearly relates to degree of heat damage, from moderate to serious. The method was developed for bread wheat, but is equally applicable to durum wheat.

Smudge and black-point

Surface discoloration due to fungi such as Alternaria alternata and Drechslera tritici-repentis pose no toxicological danger, but are serious quality defects in durum wheat. The main concern is dark specks in semolina that cause aesthetic defects in pasta (Dexter and Matsuo 1982; Dexter 1993).

In western Canada Alternaria alternata is a very common cause of “black-point”, darkening confined to the germ end, and “smudge”, a more progressed form of the infection which has spread along the crease and sides of kernels. Smudge and black-point have little effect on semolina milling yield, semolina ash, spaghetti color (except a slight trend to less brightness) or spaghetti cooking quality (Table 8). However, an increase in specks is readily apparent.

In some years red smudge, a pinkish discoloration caused by Drechslera tritici-repentis, is the predominant CWAD grading factor (Dexter 1993). Unless the infection is well progressed and induces surface darkening similar to that of black-point and smudge, quality effects are not serious, and are confined to a slight increase in semolina speckiness.

Table 8 – Effect of smudge and black-point on CWAD processing quality
Property Sample A Control Sample A Smudged Sample B Control Sample B Smudged
Smudge, % 0 3 0 3
Black-point, % 0 8 3 7
Test weight, kg/hL 83.5 83.2 83.9 83.0
Protein, % 13.3 13.5 13.3 13.5
Semolina yield, % 69.5 68.3 68.7 69.1
Ash, % 0.71 0.71 0.68 0.70
Specks per 50 cm2 48 79 55 124
     Brightness, % 45.4 45.1 47.1 44.3
     Purity, % 57.9 57.0 57.7 56.4
     Dominant wavelength, nm 578.0 578.0 577.6 577.7
Cooking score, units 12 9 12 12

Source: Dexter and Matsuo (1982). Samples of 1979 (A) and 1980 (B) crop CWAD wheat were hand-picked to give samples free of smudge and black-point (control) and enriched in smudge and black-point (smudged). Analytical data expressed on 14% moisture basis.