Measuring the potential lead contamination in the espresso machines
Many are asking me how I came up with ideas for the projects I’m working on or experimenting with. And I have to say that in most cases it is just spontaneous ideas that are emerging during article reading, video watching, or conversation. The next experiment is also influenced by a post from Barista Hustle. In their article “Wheels of Steel” they mentioned possible lead contamination coming from old brass or copper boiler, tubing, and parts, and I started researching more deeply.
In the past, there were little researches on this topic. In the year 2014, the German Federal Institute for Risk Assessment came to the conclusion that there could be a possible risk of lead leaching in the water after espresso machine descaling (1).
High levels of lead were measured right after descaling,
but the whole report is more of a guideline than thorough research. BfR is suggesting that the EU must control the materials in contact with water and must use recommended limits for lead in the water of 10ppb (parts per billion) for coffee beverages. In the mentioned study several different espresso machines were analyzed. All of them were small domestic or capsule espresso machines. No professional espresso machines were analyzed. The research I conducted aims to analyze water from professional equipment for two reasons. First, in cheap domestic espresso machines, most of the parts in contact with water are made from plastic or stainless steel, while the potential lead contamination could come from brass or copper parts used in the professional ones. The second reason is that rarely someone with a very cheap machine will put the effort into cleaning the boiler or flow paths. On the other hand, descaling is performed regularly in most coffee shops, especially those focused on coffee quality. I aim to research only professional espresso machines.
The second research on this topic was made in the year 2015 by the Ministry of Environment and Food of Denmark (2). The goal of this project was to estimate the intake of lead from coffee in the Danish population and to assess the amount contributed from coffee, water, and coffee brewers, respectively. In total 44 coffee, bean products, and 7 instant coffee products were analyzed. The analysis indicated that the content of lead in the 44 samples of coffee beans was ranging from 4.5 to 65.3 ppb with an average of 15.9 ppb. The main purpose of this research was a little different from what I was planning to do. It wasn’t about finding the source of contamination, but it only suggested that contamination could be a result of descaling.
And before you rush into a panic, I have to say that there is nothing to worry about.
In both types of research, the concentration of lead is reasonably low. Someone could say that there is no safe lead concentration but comparing it with legislation limits for a wine of 200ppb by the Commission Regulation (EC) No 1881/2006 of 19 December 2006, the levels found in coffee are lower. Still, this is a concern that should be regulated and monitored for better safety.
My project aims to compare three different scenarios related to professional espresso machines. It is not about the lead coming from the coffee as this is a topic for other research in the future. In high interest are old espresso machines produced before implementing few manufacturing improvements like lowering lead content in brass below 0.25% or coated brass.
Compare the concentration of lead in water from the heat exchanger/brewing boiler or directly from the steam boiler.
Compare the concentration of lead in water before and after the descaling/cleaning procedure.
Compare the concentration of lead in water after several cycles of purging with water after descaling/cleaning.
Results from this research will give me a possible answer to the questions below:
Where is the contamination coming from?
Is it necessary to change our descaling/cleaning procedures?
For this experiment, I collected water samples for all the mentioned scenarios and kept them in lead-free containers in a refrigerator until analysis.
I used an ICP-OES instrument for quantitation of the lead concentration in all water samples.
Inductively coupled plasma atomic emission spectroscopy (ICP-AES), also referred to as inductively coupled plasma optical emission spectrometry (ICP-OES), is an analytical technique used for the detection of chemical elements (3). Argon gas is used to create the plasma. An intense electromagnetic field is created within the coil by the high-power radiofrequency. The argon is ignited with a Tesla unit that creates a discharge arc through the argon flow to initiate the ionization process. The argon gas is ionized in the intense electromagnetic field. A stable, high-temperature plasma of about 7000-10000 K is then generated as the result of the inelastic collisions created between the neutral argon atoms and the charged particles.
An aqueous or organic sample is pulverized and introduced inside the plasma flame. The sample immediately collides with the electrons and charged ions in the plasma and is itself broken down into charged ions. The various molecules break up into their respective atoms which then lose electrons and recombine repeatedly in the plasma, giving off radiation at the characteristic wavelengths of the elements involved.
Within the optical unit, after the light is separated into its different wavelengths (colors), the light intensity is measured by an array of semiconductor photodetectors such as charge-coupled devices (CCDs). ICP-OES is widely used for the quantitation of trace elements and heavy metals like lead in concentration levels of 1-10ppb or lower. To create a calibration, I took a lead standard as Certified reference material (CRMs) with a concentration of 1000ppm and diluted it in series creating a calibration curve for the lead with concentrations of 1ppb, 5ppb, 10ppb, 20ppb, 50ppb, and 200ppb. The correlation coefficient of the calibration curve was 1.000, meaning that I got perfect linearity and reproducibility.
Note! I’m not claiming for a comprehensive or representative sample collection for this analysis. This is because of the limited number of analyzed samples and some espresso machine unknown history.
First, I measured water samples collected from the heat exchanger or brew boiler and compared them with the water from a steam boiler. It is expected that there should be differences, especially after preparing several espressos. The water in the heat exchanger or brew boiler is constantly refilling and purging with every brew, hence lowering the chance of contamination.
The water in the steam boiler, on the other hand, could spend some time heating without refilling with clean water.
This could contribute to the higher lead concentration in the water coming from the steam boiler as there is more time for lead leaching. And this is exactly what the measured samples showed with ICP-OES analysis.
On average, the lead concentration in water collected from the brew boiler or heat exchanger was below the recommended limit for lead in drinking water – under 10ppb. All measured values are summarized in Table 1:
Table 1
The values for lead vary from 0.5ppb, which roughly is the detection limit for this instrument to slightly above the recommended value of 10ppb. Here I must mention that the highest value was measured from the most recently descaled espresso machine. That makes me think about a way to test how descaling is influencing lead leaching and contamination. But this will be left for the second part of the research.
Now, let’s compare these values with the ones from the steam boiler. Few of the espresso machines showed more than 10 times higher concentration of lead in the water coming from the steam boiler and reaching values above 100ppb. According to COMMISSION REGULATION (EC), there are different limits for different drinks or food products. No international established maximum level for lead contamination in coffee or similar beverages exists, but recommended values could be encountered in some countries by the relevant regulatory authorities. These levels could vary in wide ranges and for instance, I can quote two: wine, cider, perry, and fruit wine - 200ppb; fruit juices, nectars, and concentrates - 50ppb.
Taking that into account there seems to be no serious health risk from the measured lead concentrations. Still, it raises questions and awareness.
Once again this could be explained by the longer time the water is in contact with brass and copper parts inside the steam boiler without sufficient water purging or circulation. This is a precondition for lead leaching. At a higher rate this is valid for older espresso machines where the brass could contain a higher concentration of lead, but also the welding/soldering material could contain high amounts of lead. There is a correlation between tin and lead concentration showing that the potential contamination is coming from the tin welding inside the boiler or tubes.
Many water treatment systems around the world have stopped using chlorine to disinfect drinking water. They now use disinfectants called chloramines.
Researches are showing that chloramines are associated with an increase in blood lead levels, suggesting that the use of chloramines could lead to an increase in lead exposure (5).
So, this is another potential problem for the old espresso machines. If the tap water coming from the water supply network is disinfected with chloramines, these compounds could increase the lead leaching rate.
Two conclusions could be made by this first part of the research:
Water coming from the group head could be considered safe. Descaling on the other hand could have an impact on the water quality and lead concentration.
Water from the steam boiler has a higher concentration of lead, compared with the water from the heat exchanger or brew boiler. Still, the lead contamination is at acceptable levels according to Safety Authorities but should be reconsidered to be used for other purposes than cleaning or milk steaming.
Stay tuned for the second part of this research where I'll evaluate results related to espresso machine descaling/cleaning procedures. Meanwhile, you can help me by spreading my articles. Thank you and see you soon!
(1) Freisetzung von Blei aus Kaffee- und Espressomaschinen, Aktualisierte Stellungnahme Nr. 003/2014 des BfR vom 14. Januar 2014*
(2) Exposure to lead from intake of coffee, The Danish Environmental Protection Agency, Environmental project No. 1785, 2015.
(4) COMMISSION REGULATION (EC) No 1881/2006 of 19 December 2006
(5) Environ Health Perspect. 2007 Feb; 115(2): A96.
I think I may have a more practical question for today‘S machines. Many heat exchanger machines use pure copper as boiler material so lead not as much of a concern. How much COPPER actually leaches into water/ espresso From these machines. please note that there are two sources of water, brew water for espresso which doesn’t actually sit in the boiler, ie it is drawn from reservoir and flash heated through heat exchange tube; and there is the hot water spout where water mat sit for hours in the copper boiler before being dispensed.