Achieving Flavor Parity with Alternative Sweeteners

Achieving Flavor Parity
with Alternative Sweeteners
August 1995 - Cover Story

By: Lynn A. Kuntz
Associate Editor*
*Editor since August 1996

  Fifty years ago it was difficult to find products that did not use sugar as a sweetener. Sure, some products contained traditional ingredients such as molasses or honey, but these were generally added to generate a characteristic flavor. Today, shifting markets and new ingredients have changed the sweetener systems commonly used. However, switching sweeteners can be tricky business. Often the design of the product also must change to accommodate the flavor characteristics of the alternative sweetening system.A number of ingredients, many unheard of or rarely used in commercial applications 50 years ago, are becoming increasingly common. Economics has caused a shift to corn sweeteners, particularly in liquid applications. Health and weight concerns have opened the way for other alternatives. Companies have developed new sweeteners with increasing frequency, and someday we may even get to use them in this country.

  So, food product designers must ask themselves, "How do we mimic the flavor of a sugar sweetened product? And do we really want to?

  "Each and every sweetener is different," says Gary Miller, Ph.D., vice president, research and development, McNeil Specialty Products, New Brunswick, NJ. "No matter how sophisticated a sensory test I run - and I could show that sucralose, aspartame or high-fructose corn syrup tastes virtually the same as sugar - once you integrate them into a product, each one has different flavor components. All of the products you see out there have had a significant amount of work done to balance the flavor around the sweeteners as well as other ingredients in that product."

How sweet it is

  When formulating with alternative sweeteners, the first flavor issue to address concerns the sweetness relative to sugar. This is determined by dissolving the sweetener in plain water and comparing it against a sugar-and-water standard. Analytical methods can reveal the actual level of sweetener present, but these only reflect sweetness in terms of a standard product. They will not give an accurate comparison of the sweetness level of two different products, but they can track any deterioration that occurs.

  "We use paired comparison testing," states Lisa Hanger, manager of food technology, Hoechst Food Ingredients, Somerset, NJ. "We mix up different concentrations of the sweetener and compare them to several different levels of a sucrose solution. An expert panel tastes the samples, determining which is sweeter, until they find a level that is not different statistically. It's important when making comparisons of one sweetener to another that you compare them on equivalent sucrose percentages."

  This analysis provides a best "guesstimate" of the sweeteners' relative sweetness in a food product. No aspersions are being cast on the repeatability or reliability of sensory tests; the operative phrase here is "in a food product." The relative sweetness may change depending on the matrix and the concentration. Let's just say those oft-quoted intensities may be off by a bit, depending on the exact application. They do provide a good starting point for formulation, however.

It's about time

  Tastes, including sweetness, not only produce intensity, but also exhibit temporal, or time-related characteristics. This encompasses the time of the onset of sweetness, the time of maximum sweetness perception, and the duration of the sweetness. Computerized methods have made these measurements practical, but the technique is still time-consuming and requires trained sensory panels.

  Tests show that sugar takes 4.1 seconds to reach maximum sweetness and its sweet taste lingers for 66.1 seconds. Most of the high-intensity sweeteners last approximately 10 seconds longer than sugar. However, most of the differences observed historically occur in water/sweetener solutions. As with sweetness levels, the temporal characteristics can vary with concentration and the medium. Accordingly, the product formula may need to be adjusted.

  Because of the dextrose in corn syrups and HFCS, the onset of sweetness is quicker for these than it is for sugar, but the profiles overlap. These characteristics can shift as a result of concentration or other compounds present in the product.

  "Because the sweetness of fructose comes on more quickly and dissipates more quickly than sucrose, we find it has an acid-sparing effect," notes Mike Augustine, manager of food ingredient applications, A.E. Staley Manufacturing CO., Decatur, IL. "Also, it gives you the ability to reduce flavor relative to sucrose."

  The closer the profile is to sugar, the less complicated the flavoring issue will be. When the temporal characteristics are in sync with sugar, it is more likely that a flavor system similar to one used with sugar will prove acceptable.

Beyond sweetness

  Sugar is pretty much the gold standard when it comes to flavor. In the world of sweeteners, many products trade on their claims of "tastes more like sugar." Sugar does not just provide a totally neutral sweetness; certain notes, such as a faint bitterness, can be detected by experts. The amount used in many applications can mask some flavor notes and enhance others. These nuances make an exact match elusive.

  "Sucrose initially comes on with a smooth sweet effect," says Fred Triolo, flavor chemist, creative development group at Firmenich. "It's a level effect that tails off cleanly after you swallow the product. Most of the other sweeteners have different profiles."

  One of the most popular alternatives to sugar, high-fructose corn syrup, gives a slightly different flavor. The taste is not quite as sharp due to some of the other compounds present."When tasting products made with sucrose versus HFCS, the differences are readily apparent," says Bruce Roberts, manager of beverage business development, McCormick & Wild, Hunt Valley, MD. "High-fructose corn syrup has been used so extensively in the bottled beverage industry, people don't really notice it anymore."

  The dissolution of sugar is an endothermic reaction that provides a slight cooling effect with any sugar when it dissolves. The sensation is more pronounced in certain sugars - dextrose, for example.

  Maltitol and lactitol have what is described as a pleasant sweet flavor with no aftertaste. This trait is not shared by all the polyols, however. Crystalline sorbitol exhibits a very high negative heat of solution in water. Coupled with high solubility, this gives a cooling sensation when the crystalline sorbitol is dissolved in the mouth. A similar effect occurs with xylitol. This sensation enhances mint or similar flavors.

  "Sorbitol occurs naturally in fruit, but many other components make up the flavor profile," says Lon Wilson, technical market manager, Roquette America Inc., Keokuk, IA. "With fruit flavors, sometimes it's more the acid that characterizes the flavor, not the sweetener."Some high-intensity sweeteners have what is often characterized as bitter or metallic notes. These vary with the sweetener, its concentration, the application, and the sensitivity of the person tasting it. Many attribute bitter notes to acesulfame K at high concentrations, but this effect does not appear to be universally perceived.

  "In some cases, the sweetness is so intense it is perceived as off notes," says Hanger. "The sweeteners could actually be used at lower levels. If you used sucrose at the same level, a 20% to 30% equivalency, you would find bitter notes in the sucrose itself."

  Carol Karahadian, Ph.D., manager of sensory and commercial sampling at Firmenich, observes: "If you look at chemical structures of various compounds, you would see that bitter and sweet are closely related chemically."

  Miller proposes a theory on the flavor of sucralose which also points out an inherent bitterness in sugar. A number of scientific reports show that chlorine and other halogens in certain positions on a molecule enhance the flavor components of that molecule. Sucralose is made by substituting chlorine for some hydroxyl groups of sucrose so it amplifies the taste naturally present in sugar. So, if there are any notes in addition to sweetness, they too may be amplified. One position on the sugar molecule (not one used for sucralose) actually increases bitterness if chlorine is present.

  Saccharin often gives a bitter metallic aftertaste at high levels. Using sweetener blends or adding certain compounds including tartaric acid, dipeptides and gluconates can reduce this.

Breakdown blues

  "In every system you have to be careful about compatibility and stability issues," says Bernie Ryan, group application manager, Pfizer Food Science Group, Groton, CT. "Once you develop a level of sweetness and a particular flavor, then you move on to stability. Depending on the sequence, you may have to modify or change the flavor or the sweetener system."

  We tend to think of the saccharides as eminently stable, but this is not exactly true. Under high-heat conditions, fructose, dextrose, lactose and other reducing sugars react with amino acids, forming Maillard reaction compounds. Different types of reducing sugars influence the results. Changing the sweetener system can change the notes generated through this reaction. A change from a reducing sugar to a non-reducing sugar can require the addition of the appropriate brown flavor notes in a flavor system.

  While characteristic caramelized notes can enhance many sweet products, in the wrong place, such as fruit candy, they would be considered off notes. Some of the compounds generated impart a bitter taste.

  "Reducing the pH is one way to reduce the Maillard browning," advises Augustine. "It reduces the susceptibility of the protein component. Adding something like an alkaline flow agent can create localized regions of high pH in a dry mix and result in spot browning."

  Polyols exhibit excellent thermal and chemical stability. Maillard browning and subsequent flavor development will not occur because polyols are not reducing sugars; they contain no aldehyde groups. Sorbitol resists reaction under acidic and dilute alkaline conditions.In a system that typically uses reducing sugars to generate characteristic brown notes, using non-reactive sweeteners changes the flavor profile. Birgit Mitsch, business unit manager, confectionery, pharmacy, dessert and dairy at Firmenich, suggests replacing those notes as part of the overall flavor system.

  Most high-intensity sweeteners on the market are stable in foods under normal process and storage conditions. Aspartame has several stability issues. It breaks down under high heat and low pH conditions. It can survive high heat processes such as pasteurization, however. In the neutral range, alitame is stable under the normal processing and storage conditions for bakery products and pasteurized foods. Some loss of sweetness of alitame occurs in high-heat applications such as baking, but this effect is not as pronounced as with aspartame. Ryan estimates the alitame destruction is relatively small, less than 10%.

  "In a cola at pH 3, the loss of sucralose after one year would be analytically undetectable," says Miller. "To taste a difference requires about a 10% to 15% loss. You'll get more time with the sucralose than with the cola flavor systems."

  If a sweetener is subject to loss of sweetness through process or shelf life, many formulation experts recommend increasing the initial level so any loss would bring the sweetness to the target level. This option will not work well for significant losses during long-term shelf life.

  The quandary is obvious: Where would you target the optimal sweetness, and how does that affect the other end of the spectrum? To some extent using a blend of sweeteners can overcome this problem. More stable compounds combined with those with a shorter shelf life can help ensure that an acceptable level of sweetness remains at the end of the product's shelf life.Decomposition of high-intensity sweeteners does not form any compound in sufficient concentration to affect the taste. The main effect that occurs is loss of sweetness. Benzaldehyde and cinnamaldehyde can react with aspartame in the Maillard reaction. According to Miller, sucralose is chemically inert, and the only flavor changes that occur are sensory in nature.

  "All peptide sweeteners enter into the Maillard reaction, but alitame hasn't given any results that make the product unacceptable," says Ryan. "The Maillard reaction products that are formed are so varied and at such low levels that any flavor impact is negligible."Stability leads to other formulation and flavor issues. A product using a pH-sensitive sweetener will require a different formulation and finished-product acidity than one using a sweetener that is stable under acid conditions. Sugar-sweetened colas are targeted to a pH of 2.5 to 2.75. Diet colas with aspartame tend to be around a pH of 3.2 to increase long-term stability, and they can lack the characteristic bite of the very high acid beverage.

Complex compounds

  Most food products include ingredients other than sweeteners and water, so the flavor and sweetness profile may not come across as expected. A number of factors can influence this. As with any food product, both process and ingredients affect the end product. This can further complicate flavoring issues in an alternatively sweetened product.

  "In general, no matter what sweetener you are using, you can't take one flavor and take it into a different category and expect it to taste the same," says Mitsch. "If you take a strawberry flavor in a beverage system and try to use it in yogurt, it's not going to produce the same flavor. You can make that product taste the same, but it must be achieved with a different system. The same applies to flavor systems with alternative sweeteners; they wouldn't be entirely different from those used in sucrose, but you would have to make additional changes."

  Each product category has its own challenges. With simple systems like a beverage mix, ingredients may be kept to a minimum So there is less complexity in terms of ingredient or flavor interactions with the sweetener. On the other hand, they are less likely to provide any masking or mellowing of the flavor should any problems arise.

  "Dairy products are complex and often require custom systems," says Vance Grosser, manager of dairy business development at McCormick & Wild. "Look at the various yogurts on the marketplace - the titratable acidity, the cultures they are using, the stabilizers that may be present, the heat treatment of the milk. All those factors go in to create unique situations with each. We have ongoing projects where we are trying to balance flavor, screen flavors and develop flavors, and balance the sweetener ratios.

  "We have tried to settle on one sweetener blend that will work for all flavors in one base," Grosser continues. "We've been fairly successful with that, but we've found out that when we take that same proportion to a product made with a different formulation or process, we often need to make adjustments. The other thing you may find out after you have run enough panels is that consumers may prefer certain flavors, like peach, a little sweeter, and you may need to adjust the sweeteners accordingly."

  Other tastes, as well as flavor, affect the product. Sweet is one of the four primary tastes, and each can affect the perception of the others. A highly acidic product using the same amount of sweetener as a less acidic counterpart will taste less sweet. Certain flavors, which are picked up by the olfactory system as well as the tongue, also may be perceived as sweet. Vanilla is one example of a flavor that can enhance the perception of sweetness.

  In some cases, sugar may be masking flavors that naturally occur in products. This effect may not occur when the product contains high-intensity sweeteners. Miller finds that sucralose and aspartame may enhance fruit flavors, but accentuate bitter flavors in chocolate. Using a clutched chocolate can tone down the bitterness.

  Other ingredients can alter or mask tastes and flavors. For example, starches and some gums can mask flavor, as well as sweetness. This is an important factor when using these ingredients to replace the mouthfeel of sugar or fat. The presence or absence of fat affects the overall flavor perception, also.

Double your pleasure

  When blended together, many sweeteners exhibit a synergistic sweetening effect. This usually nets an intensity increase in the 20% to 25% range. A synergistic blend also can change the profile of the sweeteners, often bringing them more in line with that of sugar.

  "What is happening in effect is that the combination is straightening out the peaks and valleys of the profiles of each individual sweetener," suggests Triolo. Synergism can make a dramatic difference in the flavor profile.

  "We do a demonstration with a nonfat yogurt base sweetened with aspartame and one using an acesulfame K/aspartame blend," Grosser relates. "Immediately you can perceive a fairly dramatic difference in the sweetness profile and the delivery. Then we have a series of different flavors to demonstrate the clean sweetness profile that you get using the combination, rather than the heavy sweetness and occasional aftertaste associated with aspartame. We aren't certain what is happening, but since you are using less of either of the sweeteners, you may have lowered that level below the threshold where any off-notes occur."

  The number of sweeteners and their varied regulatory status throughout the world make formulating for synergy a complex issue, but other synergies exist. For example, sucralose interacts with most high-intensity sweeteners, but has little synergy with sucrose or aspartame, except for modifying flavor characteristics.

  Synergy is not limited to high-intensity sweeteners. Mixtures of high-intensity sweeteners and nutritive sweeteners like fructose have a synergistic effect and help address stability issues. They can increase body and mouthfeel, as well as flavor perception. Combining fructose and sugar increases the sweetness level and may provide better flavor delivery in certain situations. If the flavor intensity of sucrose and a particular flavor peak at the same time, the sugar may create a masking effect. A blend alters the peak and allows more flavor notes to come through.

  "Fructose/sucrose blends have a synergy that is very useful in dry-mix blends," says Augustine. "Using them in a 50-50 blend will give you a relative sweetness of 128. So, without any sacrifice in sweetness you can lower the solids content. A number of factors help determine levels in a particular blend of sweeteners. It could be economics, labeling, sweetness profile, or flavor balance. We have used fructose concentrations between 1% and 5% with aspartame. In a 6-fluid-ounce beverage, that will range from 7 to 35 calories."

Flavor focus

  "One might expect that if you had flavors that work with aspartame, you could just apply them to a different high-intensity sweetener, like acesulfame K," says Grosser. "It just doesn't work that way. That's not to say that they would never work. But we ended up screening most of the flavors used with aspartame and reformulating them so they work with the acesulfame K/aspartame blends."

  All the factors reviewed here lead to one conclusion: As the sweetener system changes, the flavor system changes. The flavor system must be designed around the particular sweetener system and its characteristics. Even a switch from sugar to high-fructose corn syrup requires flavor modification to obtain a match. The other constituents in corn syrup often reduce the impact of a flavor. Several methods improve the flavor of alternatively sweetened products and bring the profile closer to that of a sugar sweetened product.

  One of the most common ways to alter the flavor is by altering the acidity or the acids used. The acid/Brix ratio is an important concept in the development of many sweetened products, especially beverages. While Brix is not the correct term for sweeteners used in the ppm range, the concept still applies.

  "Lemon is a good example. It is very sensitive to the sweetness/acid ratio," says Triolo. "If you want the product to be identifiable as lemonade it needs the right balance."

  Says Michelle Grady, Firmenich's beverage lab supervisor: "Adding a high level of acid generally lowers the perception of sweetness. Lemon and other citrus flavors and some cherry flavors often require high levels of acid."

  Recommends Roberts: "You may want to add or increase buffering agents to decrease the sharpness. You would change the flavor profile slightly - bringing some notes up, others down. We've found that products made with high-intensity sweeteners are much more sensitive to the acid level. Both the type of acid and the total acidity of the system affect the flavor perception and how the sweetening effect comes across."

  Certain acids complement certain flavors better, and the same may be true for acids and sweeteners. Several formulators report that in artificially sweetened fruit-flavored products, malic acid gives a better impact than citric acid.

  "Ingredients like citric compete with fructose because they have the same kind of onset curve. Malic acid tends to come on a little later so it will not compete with the sweetness as much," explains Augustine.

  Roberts says, "The problem is that when you get into a manufacturing situation, you frequently don't have much latitude in using different acids. From a practical standpoint, manufacturing wants to use as few ingredients as possible, so you are often stuck with a certain acid."

  If that happens, adding buffers can be one option to change the flavor perception. Using buffers with a flavor that creates a highly acidic note, such as lemon, suppresses the acidic note. However, the effect is different than if the product were formulated to the same pH without buffers. The type of buffer also can influence the final flavor. Some buffers and other ingredients can accentuate bitter or metallic notes.

  "It may help the overall flavor to look at the buffering system, the carbonation system, or the preservative system." advises Ken Patterson, product development director, The NutraSweet Co., Deerfield, IL. "There may be a flavor advantage with ingredients that are sodium-based rather than potassium-based. These won't always solve a particular problem, but they are some of the things we look at."

  The flavors themselves often must be rebalanced. For example, if a particular sweetener generates bitter flavor notes at the same time as the flavor system does, those notes may be amplified to the point of unacceptability. If a sweetener loses intensity throughout its shelf life, sometimes a flavor can be formulated to supply a perception of some of the lost sweetness.

  A number of compounds can help alter the perception of flavor and sweetness. Glycyrrhizin, a compound extracted from licorice root, and its commercial derivatives potentiate sugar and sweet flavors and help eliminate bitter notes. By itself, it is approximately 50 times sweeter than sugar. Combined with other sweeteners, it can have a synergistic effect. Because its sweetness has a delayed onset and prolonged duration, it can help mask undesirable aftertastes such as bitterness or acidity. Because glycyrrhizin possesses a strong licorice flavor, it would be unsuitable as a primary component in most products. However, the enhancing qualities occur at relatively low levels that do not generate characteristics of licorice.

  Maltol and ethyl maltol also improve the flavor and sweetness intensity. Both occur naturally in certain foods and, as commercialized ingredients, are extremely potent. Use levels in the parts-per-million range can effect a change in a finished product. They potentiate both conventional and high-intensity sweeteners, round out the flavor profile, and can mask bitterness and other off-flavors.

  "The effects are a function of application and the levels used," advises Stagnitti. "The ability of these ingredients to modify products is not one where you would look at the modification as radical but, rather, as subtleties within the formulation. It's been referred to as giving a product the final signature or finishing touches. At one level you may find you can potentiate sweetness between 10% and 15%. At other levels, you may be able to cut the duration or lingering effects of some sweeteners. We've used it effectively to mask the aftertaste of saccharin and some of the aftertaste of acesulfame K."

  Even salt can produce a positive effect when used at low levels. Its addition often can influence the overall flavor perception without a salty flavor.

  "Low levels of salt help polish the flavor. It would be easier to make an effect in a lighter system like aspartame than in one with sucrose or corn syrup. People report that in beverages they get more of a bright or sparking flavor with low levels of salt. But if you overdo it, it could cause difficulties," says Patterson.

Match point

  Most of this discussion has focused on matching sugar. However, the last point to remember is that making a product that tastes like it contains sugar may not be the right direction to take.

  "Are you trying to match something or are you trying to develop a whole new product?" asks Ryan. "Those may be two entirely different goals and require two entirely different approaches."

  In some categories, a sucrose-sweetened product may not be the best target. A large market has grown around the diet, aspartame-sweetened beverages. Many consumers have developed a preference for the flavor delivered by this system.

  "Many of the dedicated diet product users interpret the difference between aspartame and sugar as lighter and more refreshing," explains Kevin O'Malley, marketing director, The NutraSweet Co. "We all tend to follow the concept that everyone wants a zero-calorie sugar. The truth is, that might actually hurt part of the business. A zero-calorie beverage with a full-calorie mouthfeel may not always be preferred."

  Says Miller: "Initially when aspartame products were introduced, some people described the sweetness as cloying because of the lingering effect. You don't hear that any more. People have adapted to the flavor of aspartame, and it's become a new standard against which other sweeteners are compared. We've seen that kind of effect in our studies with sucralose; the more people try it, the more they seem to like the taste."

  Confectionery - in particular, chewing gum - is another category where using sweeteners other than sugar generates a flavor advantage. Patterson notes that the longer the duration of sweetness in gum, the higher the flavor rating scores.

  "In some applications you want a quick onset of sweetness; in others it may not be as important," suggests Hanger. "In chewing gum if you combine a sweetener with a quick onset and one that lingers in the mouth, you can maintain the flavor for 20 or 30 minutes."

  Working with alternative sweeteners can even help improve on nature. Most real fruit juices have a ratio of about half fructose and half dextrose. While that can be duplicated in a juice beverage, it may be possible to enhance the flavor using higher levels of fructose.

  "Generally flavors are developed and modified specifically for a sweetener system," says Stagnitti. "If you want to develop a complementary flavor system, you must bear in mind what sweetener system will be used. Flavors offer a certain amount of latitude in terms of sweetness and profile, but it's necessary to match characteristics on an individual basis. That holds true over different sweetener levels and with different flavor categories."

  Stagnitti says: "You need to formulate these products on an individual basis, keeping in mind what each component contributes to the system. It requires the systems approach. You can't change one ingredient and expect that nothing else will change. There aren't very many situations out there where that will happen."

  Whether you are matching or optimizing the flavor, working with alternative sweeteners offers a number of challenges. As time goes by, new ingredients and new markets will further complicate the task. But, after all, isn't that what we're all here for?

Back to top