Natural Sweetener Blends for Sugar Reduction: Ratios, Synergies and What the Research Actually Shows
Your reduced-sugar beverage has stevia, erythritol and monk fruit in it. You followed the supplier's suggested ratios. The tasting panel came back and said: sweeter than the full-sugar version, but it does not taste like the full-sugar version. The sweetness arrived late, stayed too long, and the product felt thin.
The problem was not the sweeteners. It was the assumption that blending is arithmetic — add the sweetness from column A, add the body from column B, and the numbers should converge on sucrose. They do not, because sucrose is not doing one job in the formula. It is doing five, and each one needs a different tool to replace it.
In this article, "natural sweeteners" refers broadly to plant-derived or naturally occurring sweetening ingredients used in reduced-sugar formulation. Their manufacturing routes differ — leaf extraction, enzymatic conversion, fermentation — and their regulatory classifications differ accordingly. The word "natural" here describes market positioning, not a single regulatory definition.
This guide covers which sweetener combinations have genuine receptor-level synergy and which ones do not, what role each ingredient actually plays in a blend, tested ratios by application with the evidence level marked for each, three industry claims that are weaker than they sound, and a five-step workflow for designing a blend that holds up in a finished product rather than only in sweetened water.
01 The problem
Why Sweetness Replacement Alone Fails in Most Reduced-Sugar Systems
Sucrose delivers five things in a finished product. Sweetness is only the first, and it is the easiest to replace.
| What Sucrose Provides | What Replaces It | Can One Sweetener Do This Alone? |
|---|---|---|
| Sweetness intensity | High-intensity sweetener (stevia, monk fruit, trilobatin) | Yes, at ppm levels |
| Temporal profile (onset, peak, decay) | Bulk sweetener with sucrose-like kinetics (erythritol, allulose) | Partially — neither matches sucrose exactly |
| Body and mouthfeel (dissolved solids) | Bulk sweetener, dietary fibre, hydrocolloid | No — high-intensity sweeteners add milligrams per litre, not grams |
| Browning and texture (Maillard, caramelisation) | Allulose (browns), erythritol (crystallises differently) | Only allulose participates in Maillard |
| Water activity and freezing-point control | Bulk sweetener or humectant | No — high-intensity sweeteners have negligible colligative effect |
Remove 10 percent sucrose from a beverage and you have taken out roughly 100 grams per litre of dissolved solids. Replace the sweetness with stevia and you have added a few hundred milligrams. The sweetness came back. The other 99 grams did not. The consumer then tastes a thin product with sweetness sitting on top of it rather than inside it, and reports that the stevia tastes artificial. As covered in our stevia aftertaste guide, the sweetener is often blamed for a problem it did not cause.
In any application where sugar contributes more than sweetness — which is most of them — a single high-intensity sweetener cannot do the job alone. That is why blending is not a formulation preference. It is a structural requirement.
The exception: partial replacement
When the target is 25 to 40 percent sugar reduction rather than full replacement, sucrose itself is still present and still carrying the body, the temporal profile and the browning. The high-intensity sweetener only needs to cover the missing sweetness. In that scenario a single stevia or monk fruit addition may be sufficient, and a multi-component blend is not always necessary. The complexity scales with the reduction target, not with the number of sweeteners a supplier can offer.
02 Synergy
What "Sweetener Synergy" Actually Means — and the Three Levels That Get Confused
Supplier literature uses the word "synergy" to describe any blend that tastes better than the components alone. That is too loose to be useful. There are three different things happening, and they are not the same.
| Level | What It Means | How It Is Measured | Example |
|---|---|---|---|
| Receptor synergy | The sweet taste receptor (TAS1R2/TAS1R3) produces a supra-additive response to the mixture — the combined signal is greater than the sum of the individual signals | Cell-based receptor assay | Reb A + erythritol (confirmed by SWEET project) |
| Sensory synergy | A human tasting panel rates the mixture as sweeter or more acceptable than the summed prediction from individual components | Trained sensory panel, psychophysical scaling | Sucralose + acesulfame K (decades of data) |
| Formulation complementarity | The blend improves the overall experience — faster onset, reduced linger, less bitterness, more body — without necessarily being sweeter | Descriptive analysis, TI or TCATA profiling, consumer acceptance | Stevia + erythritol (erythritol fills the temporal gap and adds body) |
The distinction matters because most of what the industry calls sweetener synergy is actually formulation complementarity. Complementarity is real and commercially valuable, but it does not mean the receptor is firing harder. When a supplier says "stevia and monk fruit are synergistic", the question to ask is: at which level?
The receptor-level evidence: the SWEET project
The EU-funded SWEET project published the first systematic receptor-level screen of binary sweetener blends in 2024, using cultured cells expressing the human sweet taste receptor. Six combinations were tested. The results:
| Blend | Receptor Result | Implication for Formulators |
|---|---|---|
| Reb A + erythritol | Synergy confirmed | This pairing's popularity has a receptor-level basis, not just a cost basis |
| Reb A + thaumatin | Synergy confirmed | Thaumatin does more than mask bitterness — it amplifies the sweet signal |
| Mogroside V + thaumatin | Synergy — newly discovered | Thaumatin is a synergy partner for mogroside V, not just for stevia |
| Neotame + D-allulose | Synergy — newly discovered | Synthetic + natural synergy exists at receptor level |
| Reb M + mogroside V | No synergy observed | Likely share binding sites on the receptor; 1+1 does not equal more than 2 |
| Sucralose + acesulfame K | Synergy confirmed | Classic combination re-verified at receptor level |
The authors proposed that synergy occurs when two compounds activate different binding sites on the sweet receptor, creating an allosteric enhancement. When they occupy the same site — as appears to be the case for Reb M and mogroside V — no synergy results. A separate 2024 study in Communications Chemistry confirmed that steviol glycosides bind to at least four distinct sites on the T1R2/T1R3 receptor, which explains why some steviol glycoside pairings show synergy and others do not.
The same SWEET project study also found that sucralose, acesulfame K, Reb A, mogroside V and D-allulose each activated at least two bitter taste receptors (TAS2Rs). This means that "no bitterness" is not quite accurate for any of these sweeteners — the difference is in degree, not in kind.
How to read these results
Receptor assay data describe what happens at the molecular level. They do not automatically predict what a consumer will taste in a finished product with fat, acid, aroma and protein present. A blend without receptor synergy can still be a good formulation if the complementarity is real. The receptor data tell you where to look for a free lunch and where not to expect one.
03 Roles
The Four Roles in a Sweetener Blend
Instead of asking which sweetener is best, ask which role each one is playing. The same ingredient can serve more than one role, but no ingredient serves all four.
Role 1: Intensity carrier
The molecule that provides the bulk of the perceived sweetness, used at ppm levels. This is where stevia, monk fruit and trilobatin sit. The choice between them is driven by bitterness tolerance, temporal profile and cost.
Rebaudioside A (Reb A) is the cost leader. Rebaudioside M (Reb M) has the cleanest bitterness profile but the highest cost per sucrose-equivalent point. Mogroside V monk fruit extract sits between them on cost and has a slightly fruity character. Trilobatin is the least familiar but doubles as a bitterness blocker at sub-sweet concentrations — a dual function none of the others offer. For a detailed comparison, see stevia vs monk fruit vs erythritol for beverage formulation.
Role 2: Temporal filler
The ingredient that corrects the time shape of the sweetness. High-intensity sweeteners typically start late and finish late relative to sucrose. Erythritol and allulose both have onset and decay profiles closer to sucrose, so adding them reshapes the curve even if they do not add much perceived sweetness of their own.
A TCATA study profiling 16 sweeteners at equi-sweet intensity to 10 percent sucrose found that a sucrose-allulose mixture had the temporal profile closest to sucrose of anything tested — closer than allulose alone, closer than erythritol alone, and far closer than Reb A or mogroside, both of which showed significantly delayed onset and extended bitter, metallic and chemical side-taste citations. That study also found that mogroside's temporal and side-taste profile was closer to Reb A than to sucrose, which contradicts the common positioning of monk fruit as "the cleanest" high-intensity sweetener. The difference is likely concentration-dependent: at low use levels the off-notes are below threshold, but at the concentrations needed to match 10 percent sucrose they become measurable.
Role 3: Body builder
The ingredient that replaces the dissolved solids that sugar provided. This is the role that gets forgotten most often, and it is the one that causes the "thin" or "artificial" complaint described in the stevia aftertaste guide.
Erythritol and allulose both serve here: they add volume, viscosity contribution and mouthfeel at percentage-level usage. Dietary fibres (inulin, polydextrose, soluble corn fibre) and hydrocolloids (pectin, carrageenan) can supplement further. The body builder does not need to be sweet. It needs to be present at enough grams per litre to fill the physical gap that removing sugar left behind.
Role 4: Off-note manager
The ingredient that handles bitterness, metallic notes, astringency or excessive linger — problems that the intensity carrier may introduce or that the matrix brings on its own.
Thaumatin at sub-sweet concentrations (10–50 ppm) is a broadband flavour modifier and bitterness suppressor, and the SWEET project confirmed it has receptor-level synergy with both Reb A and mogroside V. Trilobatin from sweet tea extract suppresses caffeine bitterness at 5–40 ppm according to patent sensory data, with a clear upper limit: above approximately 50 ppm in coffee the flavour itself begins to be suppressed. Mineral salt blends (KCl, MgCl₂, CaCl₂) address a different problem — lingering sweetness rather than bitterness — and have a published effect size of 5.0 to 1.6 intensity units.
Not every formulation needs this fourth role. A flavoured water at low sweetness equivalent may need only an intensity carrier and a temporal filler. A plant-protein shake with caffeine needs all four. Match the complexity to the product.
04 Ratios
Tested Ratios and Screening Ranges by Application
The table below compiles starting-point ratios from published studies, patents and supplier application data. The evidence level column tells you how much weight to put on each number. A peer-reviewed sensory study is not the same thing as a patent example or a screening inference drawn from separate data sets.
| Application | Typical Sugar-Reduction Target | Starting Blend | Evidence Level | Key Note |
|---|---|---|---|---|
| Carbonated beverage | 30–50% | Reb A 80–120 ppm + erythritol 1–2% | Patent + supplier trial | Erythritol above 1% begins to contribute sweetness and earlier onset |
| Flavoured water (zero sugar) | 100% | Stevia (Reb A or Reb M) + 3.5% allulose | Supplier sensory study (Tate & Lyle) | Sweet linger reduced; total flavour impact may also decrease — adjust flavour system. Allulose: US and approved markets only |
| Ice cream (reduced sugar) | 50–70% | Allulose 2.5–10% ± erythritol + stevia | Patent (Ingredion) | Allulose alone softens texture; pair with erythritol or LSS for firmness. Allulose: US and approved markets only |
| Protein powder (zero sugar) | 100% | Reb M 60–100 ppm + trilobatin 10–30 ppm for bitterness management | Screening inference from separate patent data | Trilobatin above 50 ppm may suppress the base flavour; validate in the actual protein matrix |
| Coffee beverage (RTD) | 100% | Trilobatin 10 ppm (bitterness blocker) + mogroside V or Reb M for sweetness | Patent sensory test (Symrise) | Trilobatin targets caffeine bitterness specifically; 10 ppm is the tested effective dose in instant coffee |
| Yoghurt (reduced sugar) | 30–50% | Sucrose 5–7% retained + Reb A 0.10–0.15 g per 100 g | Peer-reviewed TCATA study | Partial replacement; sucrose still carries body and temporal profile |
| Functional beverage with botanicals | 100% | Reb M 80–120 ppm + mogroside V + erythritol or allulose 0.5–2% + thaumatin 10–50 ppm | Supplier formulation guide (Icon Foods) | Thaumatin for adaptogen/botanical bitterness; complex system |
| Baking (full sugar replacement) | 100% | Allulose (bulk, provides browning) + Reb A or mogroside V (intensity) | Industry consensus, multiple sources | Allulose is the only natural sweetener that participates in Maillard browning. US and approved markets only |
How to use this table
These are starting points for screening, not commercial formulations. Your pH, your acid system, your flavour, your fat content, your processing temperature and your storage conditions will all shift the outcome. Ratios tested in water or model systems do not automatically transfer to finished products. Validate in your actual matrix before locking a specification. Where allulose appears, confirm that it is approved in your destination market — it is not yet authorised as a food ingredient in the EU.
05 Corrections
Three Sweetener-Blend Claims the Industry Gets Wrong
Claim 1: "Stevia and monk fruit are synergistic"
This is the most repeated claim in the natural sweetener space, and it is the one with the weakest direct support.
The SWEET project's receptor assay found no synergy for the Reb M and mogroside V blend, and proposed that both compounds may occupy the same binding sites on the sweet receptor. That does not mean the combination is useless. It means the benefit, if it exists, comes from formulation complementarity — mogroside V's mild fruit character offsetting stevia's licorice note, or their different temporal tails averaging out — rather than from the receptor firing harder. The practical difference is that you should not expect the blend to let you reduce total sweetener load. It may improve the sensory profile at the same load, but test it rather than assuming it.
For a detailed comparison of how stevia and monk fruit behave side by side, see our monk fruit vs stevia formulation guide.
Claim 2: "More erythritol solves stevia bitterness"
There is a true part and a false part here. The true part: Reb A and erythritol have confirmed receptor-level synergy, and erythritol fills the temporal gap and adds body. The false part: erythritol itself is not without sensory issues. The 16-sweetener TCATA study registered bitter, metallic and chemical side-taste citations for erythritol at equi-sweet concentrations. More noticeably, erythritol has an endothermic dissolution that produces a cooling sensation, and in beverages above roughly 3 to 5 percent that cooling becomes a new defect the panel will report.
The correct approach is to use erythritol at the level needed for body and temporal correction — typically 1 to 3 percent in beverages — and address bitterness separately through glycoside grade selection (Reb M over Reb A) or an off-note manager (thaumatin, trilobatin). Expecting erythritol to do both jobs is how cooling-sensation complaints enter the reformulation cycle.
Claim 3: "Trilobatin masks stevia aftertaste"
Patent data support trilobatin's ability to suppress caffeine and alkaloid bitterness at 5 to 40 ppm, with a sensory panel confirmation in coffee and energy drink systems. But trilobatin's effect on steviol glycoside bitterness has not been directly validated in a published sensory study. The mechanism is plausible — dihydrochalcone-class compounds have broad bitter-blocking activity — but "plausible" and "demonstrated" are different evidence levels, and this article draws the line between them.
If you are considering trilobatin for a stevia-sweetened system, test it in the target matrix at 10 to 30 ppm and evaluate whether it reduces the specific off-note your panel reports. It may work. It has not been proven to work against stevia specifically, and suppliers who claim it has are citing the caffeine data and extrapolating. For the full evidence review, see our trilobatin bitterness-blocker analysis.
06 Constraints
Cost, Regulatory and Safety Constraints That Shape Blend Decisions
Relative cost: where the budget actually goes
In most blends the high-intensity sweetener is the smaller cost line, because it is used at ppm. The bulk sweetener — erythritol, allulose or retained sugar — is used at percent levels and dominates the ingredient cost per litre or per kilogram.
| Blend Archetype | Relative Sweetener Cost per Litre | Note |
|---|---|---|
| Reb A + erythritol | Lowest | Both ingredients are the cost leaders in their categories |
| Reb A + allulose | Moderate | Allulose is more expensive than erythritol per kg |
| Reb M + erythritol | Moderate to high | Reb M is several times the cost of Reb A per SE point |
| Mogroside V 50% + erythritol | High | Mogroside V is priced above Reb A at comparable intensity |
| Reb A + erythritol + thaumatin | Moderate | Thaumatin is expensive per kg but used at 10–50 ppm |
The grade of the high-intensity sweetener matters more than most buyers realise. Reb A 60% and Reb A 97% are not the same cost per sweetness unit, and their off-note profiles differ enough to change how much off-note manager you need downstream. A cheaper starting glycoside that requires a bitterness blocker may cost more per litre than a cleaner glycoside used alone. Our stevia extract product page lists the available glycoside grades so you can compare at the specification level.
Regulatory: allulose is not available everywhere
Allulose has FDA GRAS status in the United States and an additional label advantage: it is excluded from total and added sugar declarations on US Nutrition Facts panels. It is also approved in Japan, South Korea, Mexico and several other markets.
Allulose is not authorised as a food ingredient in the European Union. It is classified as a novel food, and as of late 2026 no authorisation has been granted. If your product is sold in the EU, allulose is not a usable component in your blend, regardless of how well it performs in a US-focused formulation. Erythritol remains the primary bulk sweetener option for EU-facing natural formulations.
For a more detailed comparison of the two, including their different functional roles, see stevia vs allulose.
Safety: erythritol and the cardiovascular signal
Since 2023, a series of studies from the Cleveland Clinic group has linked higher blood erythritol levels to cardiovascular event risk and demonstrated that a 30-gram dose — roughly what a sugarless soda contains — enhances platelet reactivity in healthy volunteers within 30 minutes, with effects persisting beyond two days. A 2025 review in Cardiovascular Research characterised the state of evidence as: small pilot trials suggest temporary alteration of platelet aggregation, while studies in ICU patients given large intravenous doses and genetic studies of patients with chronically elevated erythritol do not show significant cardiovascular risk.
Three things a formulator should know. First, GRAS status is unchanged and no regulatory action has been taken. Second, no chronic dietary exposure study has been completed — this is a genuine data gap, not a settled question. Third, the relationship between dietary erythritol intake and blood levels is not linear, because the body also produces erythritol endogenously through the pentose phosphate pathway, and elevated blood levels may be a marker of metabolic status rather than a direct dietary effect.
What this changes in practice: if your brand positioning involves cardiovascular or metabolic health claims, erythritol is now an ingredient that requires a risk-benefit discussion with your regulatory and medical affairs team, regardless of GRAS status. If your brand has no health positioning, erythritol remains the most established and widely approved bulk natural sweetener, and the evidence does not currently support removing it from formulations. Monitor the research — this field is moving.
07 Workflow
A Five-Step Blend Design Workflow
This is the decision sequence that saves reformulation cycles. Work through it in order — each step narrows the field for the next one.
| Step | Question | What It Determines |
|---|---|---|
| 1 | What is the sugar-reduction target, and which label claims are non-negotiable? | Partial vs full replacement. Natural, organic, no sugar alcohols, no novel food? EU market = no allulose. Organic = confirm enzymatic Reb M with your certifier |
| 2 | Which intensity carrier fits the sensory target and the budget? | Cost-sensitive → Reb A. Sensory-priority → Reb M. Existing stevia bitterness issue → consider mogroside V, but do not expect receptor synergy with Reb M |
| 3 | Which temporal filler and body builder does the application need? | EU → erythritol. US with label advantage → allulose. Both markets → erythritol. Baking → allulose (Maillard). Keep erythritol below 3–5% in beverages to avoid cooling |
| 4 | Does the matrix require an off-note manager? | Plant protein, caffeine, botanicals, high SE → yes. Fruit-flavoured water at low SE → probably not. Trilobatin for caffeine bitterness (10–30 ppm). Thaumatin for broadband bitterness and sweetness enhancement (10–50 ppm) |
| 5 | Validate in the finished product, not in water | Build the actual formula with the actual flavour system at the actual serving temperature, then taste after the shelf life your product will really see |
The most expensive reformulation mistake is reaching for a masking agent before asking whether the intensity carrier is right for the sweetness target. The second most expensive is testing blends in water and then discovering they behave differently in a dairy or protein matrix. Both are avoidable by following this sequence rather than skipping to step 4.
We supply stevia (Reb A 60–97%, Reb D 95–98%, Reb M 95–98%), monk fruit extract (mogroside V 25–55%), erythritol, erythritol-stevia blends and sweet tea extract (trilobatin). If you send us the application, the sugar-reduction target and the destination market, we can suggest two blend architectures to screen side by side. Samples are free, MOQ is 1 kg, and quotations are returned within 24 hours. Full grade details are on the stevia extract page.
08 Questions
Frequently Asked Questions
What is the best ratio of stevia to erythritol?
There is no single ratio. In beverages, a common screening starting point is Reb A at 80 to 120 ppm with erythritol at 1 to 2 percent. Patent data show that erythritol above 1 percent begins to contribute sweetness and earlier onset, so its role shifts from pure body builder to temporal filler as the level rises. The SWEET project confirmed receptor-level synergy for this pair, which is part of why it works. But keep erythritol below 3 to 5 percent in beverages to avoid the cooling sensation becoming a defect.
Do stevia and monk fruit work better together than either alone?
They can, but not through the mechanism usually claimed. The SWEET project found no receptor-level synergy between Reb M and mogroside V — the two likely share binding sites. Any improvement from combining them comes from temporal or flavour complementarity rather than from a supra-additive sweet response. Test the blend against each component alone in your finished matrix before committing to the cost of carrying both ingredients.
Can allulose replace erythritol in a sweetener blend?
Functionally, yes in many applications. Allulose has no cooling sensation, participates in Maillard browning and has a temporal profile closer to sucrose. Regulatory is the constraint: allulose is FDA GRAS and excluded from US sugar labels, but it is not authorised in the EU, and approval status varies in other markets. If your product sells into the EU, erythritol is currently the available option.
Is erythritol safe given the recent cardiovascular studies?
GRAS status is unchanged. The 2023 and 2026 studies found associations between higher blood erythritol levels and cardiovascular markers, and a 2026 intervention study showed a 30-gram dose enhanced platelet reactivity in healthy volunteers. Limitations include small sample sizes, the fact that the body also produces erythritol endogenously, and the absence of any chronic dietary exposure study. The evidence does not currently support a blanket removal from formulations, but it does support monitoring and, for brands with cardiovascular health positioning, a risk-benefit review.
What is sweetener synergy and how is it actually measured?
Receptor synergy is measured using cell-based assays expressing the human sweet taste receptor (TAS1R2/TAS1R3). A supra-additive response — the combined signal exceeding the sum of the individual signals — indicates synergy. This is distinct from sensory synergy (panel-rated sweetness exceeding prediction) and from formulation complementarity (improved overall profile without necessarily being sweeter). The three levels are often conflated in marketing, but they describe different things.
Why does my stevia-erythritol blend taste different in protein powder than in water?
Plant proteins contribute their own bitterness and astringency, which stack with the sweetener's off-notes. Fat, pH, total solids and aroma all modulate sweetness perception. A blend optimised in water has told you almost nothing about how it will perform in a complex matrix. Build the blend in the actual product and evaluate it there.
Can trilobatin reduce stevia aftertaste?
Possibly, but the evidence is less direct than often claimed. Patent data demonstrate trilobatin's bitterness-blocking effect against caffeine at 5 to 40 ppm. Its effect on steviol glycoside bitterness specifically has not been validated in a published sensory study. The mechanism is plausible — dihydrochalcones have broad bitter-blocking properties — but plausible and demonstrated are different evidence levels. If you try it, screen at 10 to 30 ppm in the target matrix.
What natural sweetener blend works for EU markets where allulose is not approved?
Erythritol is the primary bulk sweetener for EU-facing natural formulations. Pair it with a high-intensity carrier (Reb A, Reb M, or mogroside V depending on budget and sensory target), and add thaumatin or trilobatin as an off-note manager if the matrix requires it. The Reb A + erythritol combination has receptor-level synergy, is fully approved, and has the lowest cost position of the natural options.
09 Evidence base
References
Receptor synergy and binding
[1] A receptor-based assay to study the sweet and bitter tastes of sweeteners and binary sweet blends: the SWEET project. Chemical Senses. 2024;49:bjae041. DOI: 10.1093/chemse/bjae041
[2] Steviol rebaudiosides bind to four different sites of the human sweet taste receptor (T1R2/T1R3) complex explaining confusing experiments. Communications Chemistry. 2024;7:236. DOI: 10.1038/s42004-024-01324-x
Temporal profiling
[3] Temporal sweetness and side tastes profiles of 16 sweeteners using temporal check-all-that-apply (TCATA). Food Quality and Preference. View source
[4] Temporal check-all-that-apply on the sensory profiling of sucrose-replaced sweetener blends of natural and synthetic origin. Journal of Sensory Studies. 2023;38:e12838. View source
Formulation and application data
[5] Ice cream with sucralose, stevioside, and erythritol as sugar substitutes: Sensory profile and customer preference. Food Science and Technology International. 2024;30(3). DOI: 10.1177/10820132221150534
[6] PepsiCo. Beverage and food composition with D-psicose, erythritol, and optional rebaudioside M for enhanced sweetness profile. US Patent Application. 2023.
[7] Tate & Lyle. Sensory impacts for sweetening beverages without sugar: effect of replacing stevia with allulose. Application study. View source
[8] Ingredion. Frozen dessert compositions comprising allulose and low-sugar syrup/solids. US Patent.
Trilobatin
[9] Symrise / Givaudan. Bitter alkaloid containing consumables comprising bitter blockers (trilobatin). US Patent 9,844,224. View patent
[10] Symrise. Consumables: trilobatin sweetness detection threshold and isointensity data. US Patent 10,905,140. View patent
Erythritol cardiovascular research
[11] Witkowski M, et al. The artificial sweetener erythritol and cardiovascular event risk. Nature Medicine. 2023;29:710–718. DOI: 10.1038/s41591-023-02223-9
[12] Ingestion of the non-nutritive sweetener erythritol enhances platelet reactivity and thrombosis potential in healthy volunteers. Cleveland Clinic. 2026. View source
[13] Sweeteners: erythritol, xylitol and cardiovascular risk — friend or foe? Cardiovascular Research. 2025;121(9):1319–1329. DOI: 10.1093/cvr/cvaf109
Stevia sensory and formulation
[14] Hellfritsch C, et al. Human psychometric and taste receptor responses to steviol glycosides. J. Agric. Food Chem. 2012;60(27):6782–6793. DOI: 10.1021/jf301297n
[15] Impact of different nutritive and non-nutritive sweeteners on the descriptive, temporal profile and consumer acceptance in a peach juice matrix. Foods. 2022;11(2):244. View source
Reviewed by
Xinjuan XuAssociate Professor, Ph.D. Stevia Agronomy & Crop Physiology Expert. Scientific Advisor at PhytoVale, with research focused on stevia cultivation, plant physiology, raw-material quality and steviol glycoside accumulation.
Written by
FloraContent Editor at PhytoVale. Product content and editorial specialist, focused on botanical ingredient specifications, sourcing information and B2B technical content.
B2B supply
Tell us the application, not just the sweetener
We supply stevia, monk fruit, erythritol, erythritol-stevia blends and sweet tea extract across all grade levels. Send us the product format, the sugar-reduction target and the destination market, and we will suggest two blend architectures to screen side by side. Bulk supply from 1 kg MOQ, free samples, quotation within 24 hours.
Request a sampleBrowse all products
