Image Image Image Image Image Image Image Image Image Image
Scroll to top


Using TRIZ to Make Great Cappuccino Foam

| On 07, Jul 2008

By John Cooke


Have you ever tried to make a real cappuccino drink with espresso coffee and steamed fresh milk?If you have, you know that heating the milk fully and getting the milk to foam properly is a precise art. Not everyone masters this skill and, even if you can do it, the irksome task of cleaning the steam wand and the milk jug will quickly put you off doing it too often, unless you happen to get paid to do it. What would you say if there was an easy, hassle free way to create a foamy cappuccino which was as good as the best you can get from the coffee shop? This article outlines how a number of powerful TRIZ tools were used to create a new drink system concept to deliver hassle free, great tasting and hot, foamed milk; this resulted in a patented concept that provided a unique selling point in the office coffee market and is still delivering more than $20m in incremental sales each year. (In the interest of brevity, some of the steps in the original analysis have been omitted.)

The Physics of Great Foamed Milk

As with all TRIZ analyses, it is a good idea to start by understanding the problem situation in more detail, in this case, the process of making steamed hot milk foam from the point of view of basic physics.

First, consider the bubbles. Depending on the fat level of the milk, a cappuccino foam bubble is made up of a thin skin of proteins or fats containing air. In the case of protein-stabilized milk foam, the proteins uncurl and create a thin, interconnected shell around the air bubble. The smaller the bubble, the longer the bubble will last. This is partly due to the lower surface tension experienced by small bubbles and partly due to the lower rate of drainage of water away from smaller bubbles. Small bubbles give the really creamy foam that lasts all the way to the bottom in a good cappuccino.

How do the bubbles get there in the first place? When you make a steam foamed cappuccino milk, you start with cold milk in a stainless steel jug. Next you dip the steam wand into the milk and start the steam delivery. Initially the steam is used purely to heat the milk. Once the milk is sufficiently heated and the milk is denatured to give it that cappuccino taste, you bring the steam wand close to the milk surface. This is when the bubbles really start to form as air is entrained into the milk by the steam flow and the surface of the milk is agitated by the steam. If all goes well you soon have hot milk and fine bubble foam that you can pour and spoon over the coffee to make your cappuccino. Figure 1 shows a summary of the physical actions needed to create cappuccino foam.

Figure 1: Creating Cappuccino Foam

In terms of defining the problem correctly, the next step is important and directly affects the suitability of the final solution. The precise scope of a specific problem would normally be related to business and market considerations framing the problem situation. According to the conditions of this problem, the problem scope can be defined as shown in Table 1.

Table 1: Problem Scope for Making Cappuccino Foam
Can Be ChangedCannot Be ChangedComments
Format of milk, i.e., concentration of liquid or powderMilk – constituents related to taste and foamingThis problem has been defined in terms of maximising convenience for the user based on a key consumer insight.
Steam wandThis component is the subject of the key system conflict.
AirEnvironmental resource
JugCan be replaced easily by a disposable cup, effectively resolving one harmful interaction.
Steam stateSteam useful actionsSteam can be changed to other states, e.g., water or ice, if that helps to solve the problem. Steam could also be replaced by another means provided the system is not made more complex and the consumer experience is not degraded.

The primary function of this system is purely “to foam and heat milk.” The key components in this system can be listed as follows:

Selection of Conflicting Components

  • Object:Milk
  • Main tool:Steam
  • Auxiliary tool:Steam wand
  • Foreign object/environment: Air

The term “auxiliary tool” indicates that the steam wand is helping the main tool, the steam, to perform the function of foaming milk. This identifies the best way to tackle this problem – the auxiliary tool is heavily implicated in the system conflict in this problem and so should be eliminated. This drives the system conflicts:

System Conflicts

  • System conflict 1 (SC-1): Steam wand directs the steam, which foams and heats milk, but wand becomes contaminated with milk.
  • System conflict 2 (SC-2):Absent steam wand does not direct steam to foam and heat milk, but does not become contaminated.
When the steam wand is eliminated, there is nothing left to direct the steam towards the milk. An s-field way to show this situation and the desirable final system is shown in Figure 2.

Figure 2: System Conflicts Lead to
Final System

This immediately helps to formulate a “mini-problem,” which is a statement where directionally “nothing changes but the problem goes away.” Something along the lines of: “We need to find an X-resource that will provide the useful action of directing the steam without itself becoming contaminated by milk.” The available substance-field resources include the following:

Resources of the Conflict Domain

  • Substance of the tool:”Steam”and modifications of steam
  • Energy of the tool:Kinetic energy, thermal
  • Substance of the object:Water, milk
  • Energy of the object:Thermal, chemical

Resources of the Environment

  • Substances in the environment:Air
  • Energies in the environment: Gravity, atmospheric pressure

Resources of the Overall System

  • Substances in the overall system:Jug or cup
  • Energies in the overall system: Mechanical

Normally the first resource selected to be the mystery X-resource is the tool; in this case, the “steam.” Then the mini-problem becomes an ideal final result (IFR) statement: The “steam” in the conflict domain directs itself while not becoming contaminated with milk.

Unfortunately, the “steam” as is cannot realise the IFR. This leads to the next layer of problem analysis, formulation of physical contradictions at the macro-level.

Physical Contradiction (PC) Macro

  • PC-1:In order to direct the steam, the steam must be solid/static
  • PC-2:In order to move the milk, the steam must be fluid/mobile

Elimination of Physical Contradiction Macro

  • Separation in time: This indicates a “steam” that at one time is solid/static and at another time is fluid/mobile. There are no applicable concepts from this because the two contradictory physical actions are required at the same time.
  • Separation in space: This indicates a “steam” that in one place is solid/static and at another place is fluid/mobile. A “steam” could flow where the outer layer of the “steam” directs the flow while inside the “steam” flows towards the milk surface, perhaps by changing the physical nature of the steam itself (e.g., by freezing it into a tube form – in practice quite a complex solution).
  • Separation between part and whole: This indicates a “steam” flow that is fluid/mobile as a whole but which contains components of the flow, which are solid/static in the required sense to direct the flow towards the milk surface. The steam could be directed in on itself, to direct itself. In practice this is hard to achieve with steam, but might be possible if using water flow instead. (This is a promising direction but in order to be clearer on the detail of the required steam flow, consider the physical contradiction at the micro-level.

Physical Contradiction Micro

  • PC-1: In order to direct the steam, the “steam” must consist of particles which do not move outwards from the flow direction
  • PC-2: In order to move the milk, the “steam” must consist of particles which move towards the milk surface

Elimination of Physical Contradiction Micro

Separation between part and whole: All of the “steam” consists of particles that move toward the milk surface while also consisting of particles that prevent movement outwards from the flow direction.

Consider the “steam” in different physical states. Think about the direction that began with the physical contradiction macro; consider the steam to be water. Now we need a water flow consisting of particles which move toward the milk surface while at the same time preventing themselves from moving outward from the flow. One simply way to achieve this is to use the momentum of the water molecules as they move toward the milk surface. The first conceptual model of this is a flow of fast moving water molecules surrounding slower moving water molecules, all moving toward the milk surface.

Figure 3: Model of Fast Moving
Water Molecules

In reality, all the water molecules can move quickly and so the flow velocity of the water molecules can be intensified and the momentum of the water flow employed to organize the direction in which the particles flow. Think of it in terms of the principle of a water jet; and use a jet of hot water to foam and heat the milk. In a water jet, the column of water directed toward the milk surface is held together by both its own momentum and surface tension.

Figure 4: Intensified Water Flow

With a water jet, things get better still when considering the required function “moves air.” Due to the boundary layer effect, air molecules directly contacting the outer surface of the jet will have the same velocity as the jet and will be entrained into the milk with the jet. At this point it is a good idea to check that the water jet idea still delivers all the required useful actions of the steam (Table 2).

Table 2: Water Jet Principle to Move Water
Steam Useful ActionsDoes Water Jet Deliver?Comments
Moves airYesA water jet is an effective way to entrain air and has been used in the sewage treatment industry for many years.
Moves milkYesDepending on the jet velocity and milk characteristics, a water jet can create different bubble sizes in an easily controlled manner.
Heats milkYesThis action can be delivered by using a concentrated milk format and using a hot water jet to dilute the milk concentrate. This is the final step in the conceptual solution and incidentally plays to the advantage of the host drinks system which already uses powder concentrates to make hot drinks and has an integral hot water delivery system.


The final design implementation made use of a high velocity hot water jet delivery system integrated into an upgraded office coffee brewing system, delivering ease of use and high quality coffee shop style drinks which 74 percent of consumers stated were as good as or better than those they could get from Starbucks. The concentrated milk is provided as a consumable with the drinks system and is sealed in individual packs dispensed at the point of preparation. The final system is incredibly easy to use compared to using normal milk and does not suffer from any of the hygiene issues associated with fresh milk. There is now a “virtual” steam wand, created by the water jet, diluting the concentrated milk and foaming it to create milk with the desired creaminess and texture of authentic hot foamed cappuccino milk.

The solution seems simple, but in practice, the analysis was iterative and required a significant amount of research and supportive experimentation. Nevertheless, use of TRIZ tools significantly influenced the outcome of a product development, reducing implementation time significantly and creating a patent-protected solution that is still in the office coffee market, generating more than $20 million of incremental sales every year and providing a sustainable unique selling point. Most importantly, this TRIZ solution provides consumers in the office with a much more convenient alternative to the coffee shops on every block.


Particular thanks to Victor Fey for his mentoring and support in TRIZ. Victor’s insightful training has really helped me to get the most from TRIZ.


  1. Fey, Victor and Rivin, Eugene, (2005) Innovation on Demand: New Product Development Using TRIZ, Cambridge University Press.
  2. Rouimi, Sandrine, Schorsch, Catherine, Valentini, Céline and Vaslin, Sophie, “Foam Stability and Interfacial Properties of Milk Protein–surfactant Systems,” Food Hydrocolloids, Volume 19, Issue 3, May 2005, Pages 467-478.
  3. Saint-Jalmes, A. , Peugeot, M.-L., Ferraz, H. and Langevin, D. , “Differences Between Protein and Surfactant Foams: Microscopic Properties, Stability and Coarsening. Colloids and Surfaces,” Physicochemical and Engineering Aspects, Volume 263, Issues 1-3, 1 August 2005, Pages 219-225.
  4. “Surface Entrainment of Air by High Velocity Water Jets, Chemical Engineering Science, Volume 29, Issue 4, April 1974, Page 1056.