Molecular Weight Distribution in Polymers: Averages, Dispersity, and How to Calculate Them

Why Polymers Have a Distribution, Not a Single Molecular Weight

Molecular weight and molar mass are two terms that indicate how large a polymer chain is. Although used interchangeably, molecular weight is in practice the term used most throughout the polymer industry, whilst molar mass is the IUPAC-preferred term for its dimensional precision, expressed in g/mol.

What makes polymers unusual is that a single molecular weight value can never fully describe them. Unlike a small molecule, where every individual molecule is chemically identical, a polymer sample contains chains of many different lengths – and therefore many different molecular weights – all coexisting in the same material. This spread of chain lengths is the molecular weight distribution (MWD) (also referred to as molar mass distribution (MMD)), and characterizing it, rather than reducing it to a single average, is the starting point for understanding what a polymer actually is.

Chain-Length Variation as an Intrinsic Feature of Polymerization

During polymerization, individual chains do not grow to a pre-set length and stop. Initiation, propagation, and termination events happen probabilistically: a chain may terminate after a few hundred monomer additions or after tens of thousands, depending on when a termination event occurs relative to that particular chain. The result is that any commercial polymer contains a broad mixture of chain lengths, and therefore a broad mixture of molecular weights.

How Polymerization Mechanism Determines Distribution Shape

The shape of the MWD is not random, it is largely predictable from the polymerization route. Three commercially significant examples illustrate this:

 
Metallocene catalysis (single-site catalysts used for mLLDPE, metallocene PP, and ethylene/α-olefin copolymers) provides a single, well-defined active site. Every chain grows under essentially identical conditions, producing a narrow, near-symmetric MWD. Typical dispersity (Ð) values are close to 2. The narrowness that makes metallocene polyolefins attractive for optical and mechanical performance also makes them less tolerant of process variation, as there is less breadth in the distribution to absorb fluctuations.

Polyethylene copolymers are the simplest structures produced in single-site catalysts

Comonomer is incorporated into the linear Polyethylene Chains

 
Ziegler-Natta catalysis (the basis for most commodity HDPE, LLDPE, and isotactic PP) uses heterogeneous catalysts with multiple types of active sites. Each site type produces chains with a different average molecular weight. The combined output is a broad, often multimodal distribution, and typical Ð values of 4–8 or higher are normal for these grades. If your GPC/SEC chromatogram shows a narrow MWD for a material you know to be Ziegler-Natta PP, that is worth investigating.

Resins where non-uniform comonomer is added in multi-site catalysts

Comonomer added in multisite catalysts is not homogeneous

High-pressure radical process (used for LDPE, and ethylene copolymers like EVA and acrylate-modified PE) involves chain transfer and long-chain branching reactions that generate an exceptionally broad distribution, where Ð values of 8–20 are typical. This breadth, combined with long-chain branching, gives LDPE its characteristic melt strength and bubble stability in blown film.

Describing the Distribution: The Key Molecular Weight Averages (Mn, Mw, and Mz)

The full molecular weight distribution is a curve, but most routine communication relies on three derived numbers: Mn, Mw, and Mz. Each is a different kind of average, sensitive to a different part of the distribution. The key is understanding which question each one answers, because in polymer science, different physical properties – such as tensile strength, entanglement behavior, melt elasticity, melt viscosity, and brittleness – correlate to different statistical “weights” within the distribution.

A typical MMD curve of three different resins

 
Number-Average Molecular Weight (Mn): Definition, Formula, and What It Reflects
Mn answers the following question:

What is the average molar mass if every chain in the sample counts equally, regardless of its size?

Mn = Σ(Nᵢ · Mᵢ) / Σ(Nᵢ)
where Nᵢ is the number of chains with molar mass Mᵢ. In plain language: count every chain, add up its molar mass, divide by the total number of chains. Because short chains are more numerous than long chains in a typical polymer sample, Mn is pulled toward the lower end of the distribution.

The practical consequence is that Mn is the molar mass average most sensitive to the low-mass tail of the distribution. A modest population of very short chains – even if they represent only a small fraction of the total sample mass – will drag Mn down noticeably. In polyolefins, short chains below the entanglement threshold are associated with presence of oligomers, extractables, migration potential in food-contact applications, and a reduction in mechanical properties like impact resistance and ductility.

Weight-Average Molecular Weight (Mw): Definition, Formula, and Why It Differs from Mn
Mw answers the following question:

What is the average molar mass if each chain is weighted by its contribution to the total mass of the sample?

Mw = Σ(Nᵢ · Mᵢ²) / Σ(Nᵢ · Mᵢ)
A long chain contributes far more mass to the sample than a short chain, so it receives proportionally more weight in this average. Mw is therefore always greater than or equal to Mn (equal only in the hypothetical case of a perfectly monodisperse sample where every chain is identical).

Mw is the molar mass average that best predicts bulk mechanical properties and melt rheology. Melt viscosity, tensile strength, and resistance to slow crack growth all correlate strongly with Mw. In the manufacturing of pipe-grade HDPE or a film-grade LLDPE, Mw is typically the most important average from a GPC/SEC run.

Z-Average Molecular Weight (Mz): When and Why It Matters
Mz is a higher-order average that gives even greater weight to the largest chains in the distribution than Mw does.
Mz = Σ(Nᵢ · Mᵢ³) / Σ(Nᵢ · Mᵢ²)
Mz is directly relevant to melt processing behavior. Very long chains drive the elastic component of melt flow: die swell, parison sag resistance in blow molding, and susceptibility to flow-induced crystallization all correlate with Mz. In HDPE blow molding grades, a drop in Mz that leaves Mw unchanged can manifest as parison instability long before any standard mechanical test would detect a problem. When processing consistency is critical and Mn and Mw look normal, Mz variation is often the culprit worth checking.

Example: Calculating Mn, Mw, and Dispersity from a Discrete Distribution

The table below shows a hypothetical five-fraction distribution. Although real GPC/SEC analysis operates on a continuous distribution – integrating across the full elution curve rather than summing discrete fractions – the arithmetic principle is identical.

From the table below:
Mn = 10,000,000 / 200 = 50,000 g/mol
Mw = 1,460,000,000,000 / 10,000,000 = 146,000 g/mol
Ð = 146,000 / 50,000 = 2.92.

Additional Molecular Weight Averages: Mz+1 and Mp

Mn, Mw, and Mz cover the most used statistical moments of the molecular weight distribution. Two additional descriptors – Mz+1 and Mp – appear less frequently in routine reporting, but they can be useful in certain scenarios as they carry specific diagnostic value that the three standard averages cannot provide.

Z+1 Average Molecular Weight (Mz+1): Sensitivity to the Extreme High-Mass Tail

Mz+1 extends the hierarchy of molar mass averages one step further toward the extreme high-molar-mass end of the distribution. Its formula is:
Mz+1 = Σ(Nᵢ · Mᵢ⁴) / Σ(Nᵢ · Mᵢ³)
Mz+1 is the most sensitive to the extreme tail of the distribution.
For most routine analyses, Mz is sufficient. The z+1 average becomes relevant when the extreme tail is suspected to be driving a specific problem.
Reliable Mz+1 values require strong detector signal at the high-molar-mass end of the chromatogram, precisely the region where polymer concentration is lowest and baseline noise has the greatest relative impact.

Peak Molecular Weight (Mp): The Most Probable Chain Length

Mp captures the most probable chain length directly from the chromatogram peak, independent of tail behavior. It is not a calculated average, it is read directly from the chromatogram: the molar mass value corresponding to the highest point on the MWD curve.

The usefulness of Mp as a diagnostic value depends on the breadth and type of distribution:

• In narrow monomodal grades, Mp is sharp, reproducible, and sensitive to small process shifts. In these grades, Mp can be useful for tracking batch-to-batch consistency of the peak population.
• In broad Ziegler-Natta grades, the peak of the weight-fraction curve is flatter. Therefore, the Mp value has higher uncertainty and should be interpreted cautiously alongside the full chromatogram rather than as a standalone number.
• In bimodal distributions, a single Mp value is not meaningful, and deconvolution into two separate Mp values would be necessary for this parameter to carry any diagnostic value.

Dispersity (Ð = Mw/Mn): The Single Most Informative Ratio

If you need to summarize an entire molecular weight distribution in one number, dispersity is that number, but it is important to understand exactly what it captures.

 
What Ð Tells You About the Polymerization Process
Ð = Mw / Mn

The symbol Ð (eth) is the IUPAC-recommended notation introduced in 2009, replacing the older term polydispersity index (PDI), although both terms remain in wide use.

Ð = 1.0 is the theoretical limit for a perfectly monodisperse polymer, where every chain is identical in length and molar mass. This is never achieved for synthetic polymers in practice. In production, Ð is a rapid first indicator of polymerization control: a narrow Ð (approaching 2 or below) points toward a single-site or living polymerization; a broad Ð (4 or above) points toward a heterogeneous catalyst, a reactor blend, or a free-radical process.

That said, Ð compresses a complex distribution into a single number, and compression always loses information. Two polymers that have identical Mn, Mw, and Ð can still have very different distribution shapes. These materials will process and perform differently in end use, this is why examining the full chromatogram matters.

Why Averages and Dispersity Alone Is Not Sufficient – The Shape of the Molecular Weight Distribution Matters

Consider three HDPE samples with identical Mn, Mw, and Ð: one with a symmetric MWD centered on Mw, one with a pronounced high-molecular-weight tail (increased melt elasticity, tendency toward die swell, gel risk in film), and one with a low-molecular-weight shoulder (potential extractables, reduced impact resistance). All three reporting the same Ð. They would not, however, process the same way or deliver the same end-use performance.

This is why the full MWD chromatogram – not just the derived averages – should always be examined when diagnosing a process upset, comparing candidate grades, or qualifying a new material supplier.

Three HDPE with the same Mn, Mw, and Dispersity but with different MWD, which means different processing behavior and performance

Distribution Shapes in Polymers and What They Mean

Learning to read a GPC/SEC chromatogram qualitatively is one of the most practically valuable skills in polymer characterization. The peak position tells you where the mass of the MWD sits, and the shape tells you about the process that made it and the performance you can expect.

Monomodal Distributions: Narrow vs. Broad

A monomodal distribution has a single peak. The critical distinction is between narrow (Ð close to 2) and broad (Ð of 4 or above). On a standard log(M) vs. normalized weight fraction plot – the default output of GPC/SEC software – narrow MWDs appear tall and sharp, while broad MWDs are low and wide.

Narrow monomodal distributions are the signature of metallocene polyolefins. The uniform chain population delivers consistent properties and makes rheological modelling more predictable. The trade-off is that a narrow MWD also means a narrower processing window. Small changes in Mn or Mw have proportionally larger effects on viscosity and mechanical performance because there is less breadth in the distribution to absorb them.

Narrow Molar Mass Distribution curve tyipical of Metallocene Catalysts

Broad monomodal distributions are the norm for Ziegler-Natta catalysis. The wide range of chain molecular weights provides some intrinsic processing latitude (lower-molecular-weight fractions act as a natural lubricant during extrusion, reducing torque demand) but also introduces higher variability in end-use properties across production batches.

Broad Molar Mass Distribution curve tyipical of Ziegler-Natta Catalysts

Bimodal and Multimodal Distributions: How and Why They Arise

A bimodal distribution shows two distinct peaks, or a pronounced shoulder that resolves into a second peak with adequate column resolution. In many commercial polyolefins, this is not an anomaly, it is an engineered feature.

In a dual-reactor (cascade) process, two distinct polymer populations are produced and combined at the reactor level: a high-molecular-weight component providing mechanical performance and a low-molecular-weight component providing processability. Pipe-grade HDPE – also known as PE100 and its higher grade PE100-RC (Resistance to Crack) – is the most prominent example: the bimodal MWD is carefully balanced so that the high-molecular-weight population resists slow crack growth over decades of service while the low-molecular-weight population keeps melt viscosity within a processable window.

Bimodality matters in both directions. If a chromatogram shows bimodality in a material that should be monomodal, it may indicate contamination, a reactor upset, or an unexpected blend. Conversely, if a bimodal product appears monomodal, the two populations may have merged, which could be a sign of a process problem.

Bimodal Molar Mass Distribution curve

Skewed Distributions: High and Low-Molecular-Weight Tails and Their Practical Consequences

Few real polymer MWDs are perfectly symmetric. The tails – the extended edges of the distribution at high and low molecular weight – carry information that Ð alone cannot convey.

A high-molecular-weight tail (positive skew) significantly increases Mz and drives elastic behaviour in the melt. In LDPE, the long-chain branching that creates the high-molecular-weight tail is exactly what gives the material its melt strength and bubble stability in blown film processing. In HDPE, an unexpected high-molecular-weight tail can cause gel formation and surface defects in film applications.

A low-molecular-weight tail (negative skew) introduces short chains that fall below the entanglement threshold. In small amounts, these chains improve processability by reducing melt viscosity. In excess, they reduce impact resistance, increase extractable content, and create migration risk in food-contact packaging applications.

Polyolefin Case Studies: HDPE Pipe, LLDPE Film, PP Impact Copolymers

HDPE pressure pipe (PE100 and PE100-RC grades): Bimodal MWD by design. The high-molecular-weight component resists slow crack growth, and the low-molecular-weight component keeps melt viscosity within a processable window for large-diameter pipe extrusion. A shift in Mn specifications – indicating a change in the low-molecular-weight population – or a change in the bimodal peak ratio can predict long-term performance issues years before they would appear in any mechanical end-use test.

LLDPE blown film: Metallocene grades (Ð ≈ 2) are characterized for having excellent optical properties and outstanding puncture resistance, but are sensitive to process conditions, requiring tighter control of melt temperature and screw design, often requiring them to be blended with LDPE to improve bubble stability. Conventional Ziegler-Natta LLDPE (Ð ≈ 4–5) is easier to process on standard equipment but lacks mechanical performance for this application (inferior dart impact, puncture, and optics). The MWD is the fundamental differentiator between these two commercial families, and GPC/SEC is the primary tool for verifying this.

PP impact copolymer (heterophasic PP): These grades contain a PP matrix phase and a dispersed ethylene-propylene rubber (EPR) phase. The overall MWD reflects contributions from both phases simultaneously. A shift in the high-molecular-weight region of the chromatogram often signals a change in the rubber phase (its content, molecular weight, or both) which directly affects low-temperature impact performance. High-temperature GPC with IR detection (Polymer Char GPC-IR) can differentiate PP and ethylene-containing fractions by their IR spectral signatures, providing composition information for each molecular-weight eluting fraction, information that a standard refractive index detector cannot deliver.

Related Article: High-Temperature GPC: A Complete Guide 

Measuring Molecular Weight Distribution: GPC/SEC Technique

Gel Permeation Chromatography (GPC) – also known as Size Exclusion Chromatography (SEC) – is the standard technique for MWD measurement in polymer laboratories.

GPC/SEC separates polymer chains by their hydrodynamic volume (the effective volume a chain occupies in solution) not by molecular weight or molar mass directly. Converting this separation into a molar mass distribution requires a calibration curve that relates elution volume to molar mass, and that calibration relies on a critical assumption: that there is a predictable, consistent relationship between hydrodynamic volume and molar mass for the polymer under measurement.

Conventional Calibration vs. Universal Calibration vs. Absolute Methods

Conventional calibration uses narrow-distribution polymer standards to construct the calibration curve. Molar mass values for unknown samples are read directly from this curve. Molar mass values are only correct if the polymer under analysis has known Mark-Houwink constants and these constants do not change along its molar mass distribution.

Universal calibration uses the product [η]·M (intrinsic viscosity multiplied by molar mass, proportional to hydrodynamic volume) to construct a calibration that is, in principle, independent of polymer type. Requires either an online viscometer detector or known Mark-Houwink constants for the specific polymer-solvent-temperature system. More accurate than conventional calibration when crossing polymer types, but still model-dependent.

Absolute methods (MALS + RI/IR) Multi-angle light scattering directly measures the weight-average molar mass of each eluting fraction without assuming any relationship between hydrodynamic volume and molar mass. Essential for any application where the absolute accuracy of Mw values matters.

READ MORE: Introduction to GPC/SEC

1. Flory, P.J. (1953). Principles of Polymer Chemistry. Cornell University Press. Foundational treatment of molecular weight distribution statistics and polymerization kinetics.
2. Striegel, A.M., Yau, W.W., Kirkland, J.J., Bly, D.D. (2009). Modern Size-Exclusion Liquid Chromatography, 2nd ed. Wiley. Standard reference for GPC/SEC theory and practice.
3. IUPAC (2009). Recommendations on polymer nomenclature: source for the term dispersity (Ð) and associated notation. Pure and Applied Chemistry, 81(2), 351–353.
4. Spalding, M. A., & Chatterjee, A. M. (Eds.). (2017). Handbook of industrial polyethylene and technology: Definitive guide to manufacturing, properties, processing, applications and markets. John Wiley & Sons.
5. Polymer Char Application Notes and Publications Library. HT-GPC and GPC-IR methodology for polyolefin characterisation. Available at: polymerchar.com/library/publications