Project Summary

Determination of Aromatic Production from Surface Browning to Improve Flavor in Steaks Using Differences in Steak Thickness and Cook Surface Temperature

Principle Investigator(s):
C. R. Kerth, Ph.D.
Texas A&M University
Completion Date:
May 2013



Beef flavor is an incredibly complex and completely understood quality trait. While Calkins and Hodgen (2007) and Mottram (1998) have done an admirable job of reviewing current literature and describing many of the factors involved in determining beef flavor, much still needs to be done. Basic flavor components are generally classified into two categories: lipid-derived and Maillard reaction products (MRP). During cooking the lipids are degraded giving various derivative compounds that are aromatic, but have much higher aroma thresholds (higher concentrations are required to detect an aroma) compared to MRP. On the other hand, MRP are a result of the combination of heat, sugar, and amino acids. The meat supplies the sugar (primarily ribose) and amino acids from the high protein content, so when meat is heated to over 300°F at the surface, the result is browning and the production of hundreds of aromatic compounds with very low aroma thresholds (parts per quadrillion in some cases). Specific objectives were 1) to determine the primary groupings of chemical aromatics that generate positive beef flavor ID’s using trained panel and GC/MS data generated from the Beef Flavor Lexicon experiment (R.K. Miller and C.R. Kerth, NCBA, 2011) 2) to determine the ability to generate and document positive aromatic chemical compounds generated by steaks cut 0.5, 1.0, or 1.5 inches thick cooked to a medium degree of doneness at a low (350°F), medium (400°F), or high (450°F) grill surface temperature.  


Steaks. Select strip loins were ordered from a commercial distributor from a major packing plant. The strip loins were placed in the freezer after aging to allow uniform and precise cutting thickness of the steaks on the band saw. After intact strip loins were frozen, loins were divided into 9 portions (3 cooking temps x 3 steak thicknesses). Steaks for all cooked analyses were placed on the grill, turned when the internal temperature reaches 35°C (95°F) and removed when the internal temperature reaches 71°C (160°F, medium degree of doneness).

350°F 400°F 450°F
0.5” 1.0” 1.5” 0.5” 1.0” 1.5” 0.5” 1.0” 1.5”
≥ 6 steaks ≥ 6 steaks ≥ 6 steaks ≥ 6 steaks ≥ 6 steaks ≥ 6 steaks ≥ 6 steaks ≥ 6 steaks ≥ 6 steaks

  • (1 strip loin cut into 9 steaks representing all 3 thicknesses = 9 inches, sufficient to square rib end and avoid vein steaks)

GC/MS and Sniff-port. Steaks were cooked, all external fat were removed, and the steaks were cut into pieces as would be done for sensory (0.5 in x 0.5 in cubes) and placed in a 1 pint glass jar with a Teflon lid and placed in a water bath held at 60°C (140°F). After equilibrating for 20 minutes, a solid-phase micro-extraction (SPME) fiber was inserted through the lid and into the headspace above the steak to collect volatile chemicals.


To determine the impact of steak thickness and cooking surface temperature, the volatile compounds that are associated with published aroma descriptors are reported in Table 1. The highest cook surface temperature (450°F) produced higher amounts of 1-Hexanol (cut grass), 2-ethyl-5-methyl-pyraine (roasted), 2,5-dimethyl-pyrazine (roasted), phenyl-acetaldehyde (rosy), and trimethyl-pyrazine (roasted), but lower total amounts of 2,3-butanedione (buttery), and toluene (paint thinner) compared to the lower cook surface temperature. The thinnest steaks (0.5 inch) dominated those volatiles primarily originating from lipid degradation from the shorter cooking times. These included those that have a tallow, woody, soapy, fruity, greenbean, smokey, fruity, oily, green, fatty, and sweet aromas from E-2-decenal, 1-hexanol, 1-octanol, 2-heptanone, 2-pentyl-furan, decanal, ethyl-benzene, heptanal, nonanal, octanal, styrene, respectively. On the other hand, the 3.81-cm-thick steaks dominated the production of Maillard products with roasted, roasted, roasted, fruity, burnt, buttery, roasted, meaty/roasted, sickly, and roasted aromas generated by 2-ethyl-3-methyl-pyrazine, 2-ethyl-3,5-dimethyl-pyrazine, 2-ethyl-5-methyl-pyrazine, 2-ethyl-6-methyl-pyrazine, 2-methyl butanal, butanedione, 2,5-dimethyl-pyrazine, 3-methyl-butanal, and methyl-pyrazine, respectively.

1-Pentanol and 2-decenal (lipid degradation products) variation both could be explained by steak internal temperature on and off with over 22% and 12% of the variation, respectively. About two-thirds of the variation in 2,4-nonadienal (fried, fat aroma) could be explained by skillet temperature at the beginning and end, steak surface temperature at the flip and at the end, steak internal temperature at the end of cooking, and cook loss percentage. Butanoic (butyric) acid is characterized by a very strong, unpleasant aroma similar to vomit at high concentrations, but has been shown to be important to aroma balance at low concentrations, and five cooking measurements accounted for 17.0% of the variation. Almost 20% of the variation in the smokey aroma decanal can be explained by beginning, flipped, and ending steak surface temperature, total cooking time, and cooking loss percentage. Twenty-four percent of the meaty aroma 2-methyl-furan generated from Maillard reactions is solely from the steak surface temperature at the end of cooking. Beginning steak internal temperature and ending steak internal temperature accounts for 35.8% of the variation in the green, burnt aroma of heptanal. Hexanal (an extraordinarily important volatile compound that indicates the degree of oxidative rancidity/warmed-over-flavor) and pentanal (a green, burnt aroma) are both lipid degradation products and 17.8% and 9.6% of the variation, respectively, can be attributed to the ending internal steak temperature. A total of seven variables entered the equation for nonanal (green) and accounted for almost 29% of the variation. Four temperature variables combined to account for more than 35% of the variation in the fatty aroma from octanal. The three Maillard products 2,3-dimethyl-pyrazine, 2,5-dimethyl-pyrazine, and trimethyl-pyrazine with nutty, roasted aromas could be described by the ending surface temperature, and when combined with other cooking temperature and time measurements accounted for 22.2, 49.5, and 21.4% of the variation, respectively. Additionally, all three of these compounds were positively affected by both time and temperature which would be characteristic of Maillard products.


Steaks can be cut to different thicknesses and cooked with differing cook surface temperatures to differentially create aromatic volatiles that are characteristic of various aroma descriptors. As we find more information about the likes and dislikes of consumers with differing backgrounds (e.g. heavy, moderate, or light beef users) and desires, it appears that it will be possible to give specific cutting and cookery instructions to generate volatile aromatic compounds that match those consumers’ likes. This will become increasingly important as beef consumers as a whole have more and more differing perceptions and acceptances and the beef industry is able to create a product that is liked by any beef consumer, especially in the less-expensive commodity grades of USDA Select and Low Choice.  

Table 1. Cooking surface temperature and steak thickness impacts on volatile compounds with published aroma descriptors.

Header Header Header Header Header Header
Header text text text text text
Header text text text text text
Header text text text text text
Header text text text text text
Header text text text text text
Header text text text text text
Footer text text text text text

  • a,bMeans in a row within either the temperature or steak thickness treatment with different superscripts differ (P < 0.05)

Figure 1. Cooking surface temperature affects skillet and steak temperatures.

Figure 2. Steak thickness affects skillet and steak temperatures.