Please romaine calm- there might be cancer-causing compounds in your lettuce

Lee, Wan-ning, Huang, Ching-hua, and Zhu, Guangxuan. 2018. Food Chemistry. Analysis of 40 conventional and emerging disinfection by-products in fresh-cut produce wash water by modified EPA methods. Food Chemistry (256):319-326.


From the Fields to Your Countertops

Think back to your last trip to the grocery store. All the produce is sorted and stacked nicely and tends to look perfect (no ugly fruit here). Even more than that, however, is that it’s clean.

ugly strawberry

It’s not the strawberry’s fault it’s ugly. Source: Flickr.

Not clean as in you shouldn’t wash it when you start cooking- definitely give it a good scrub when you get home! – but certainly not caked in soil or fertilizer residue. All produce goes through some kind of wash process prior to arriving to a supermarket (not just the “pre-washed” stuff).

Flume washer

This is the part that I like to think of as a lazy river for plants. Source: leafy vegetables processing line, Sormac.

In addition to removing soil, the washing process is designed to remove bacteria which may help prevent outbreaks, such as the recent romaine lettuce-E. coli outbreak in the United States. Although this outbreak wasn’t attributed to any issues with the washing process, it gives us an idea of the impact and scale of contaminated produce: according to the US Center for Disease Control, so far, approximately 200 people were infected in 35 states, 89 were hospitalized, and five died from consuming these contaminated leafy greens.

At left, a CDC map published June 1, 2018, of the multi-state romaine outbreak in the United States. At right, its innocent-looking perpetrator. Sources: CDC, Pexels.

Chlorine (commonly as sodium hypochlorite) is the standard disinfectant for washing produce as well as for treating drinking water.  But as you might imagine, there’s a big difference between treating filtered, clean drinking water and treating fruits and vegetables.

Taking a Load Off

The difference is organic load. Think about taking a bath: when you soak in a bathtub, dead skin, dirt, and oils end up in the water. The same goes for soaking produce! The technical definition of organic load is the amount of particulates, both solid and dissolved, released into water (learn more about organic load in drinking water related to storms in another envirobites article here). The higher the organic load resulting from the produce, the more work the chlorine treatment needs to do. In short, to get the same bacteria-killing power, you probably need to add more chlorine.

There’s a danger, though, in adding more chlorine. Chemicals both in the produce and coming off the produce can be chlorinated- so the chlorine atoms pop off of the disinfectant and onto the chemicals in the produce and in the water.

The effects of this are two-fold: again, your chlorine dose is effectively lowered because it hangs on to “dirt”, so you need to add more chlorine to get the same bacteria-killing power, and second: these chlorinated compounds tend to be carcinogenic.

An olive-washing machine. Chlorine content unknown. Source: Max Pixel.

The increased cancer risk from these compounds, called “disinfection by-products” or DBPs, is estimated to be a small amount over a person’s lifetime, and the United States Environmental Protection Agency set maximum limits on the acceptable limits of many of these compounds in drinking water. However, these limits don’t include all DBPs- especially the ones more recently discovered.  And despite the extensive use of chlorine in free produce washing, there is almost no data on DBPs in the leftover wash water or formed in the produce itself.

A Problem with Tools

Part of the issue in research is measurement– we don’t currently have the right techniques to measure DBP levels in wash water or in produce reliably. There are some ideas out there, but as Wan-Ning Lee, first author and lead graduate student on this study says, “You can count all the studies on this on two hands.”

It’s in part a question of measuring power- how sensitive our instruments are- as well as just how complicated these materials are. It’s easy to imagine how complicated fruits and veggies are- looking for small amounts of a certain DBP in a whole strawberry, which is made up of lots of chemicals, can be like looking for a needle in a haystack.

With the leftover wash water, the issue goes back to the issue of organic load. Having lots of chemicals in the water makes it challenging to correctly identify and measure specific, individual DBPs. While there are existing EPA methods available for measuring DBPs in drinking water, those methods don’t work well for the produce wash water.

Lee recently saw this problem and got to work, publishing a paper in Food Chemistry in February. She first focused on the wash water, not the produce itself- though she assures me that’s coming- and studied 40 different regulated and emerging DBPs, trying to improve the sensitivity and reliability of the instruments.

Not all chemistry equipment is high-tech! At left, strawberries spinning; at right, lettuce soaks. Source: Wan-ning Lee.

Don’t Play with Your Food

To conduct this study, Lee went grocery shopping! She bought fresh strawberries, lettuce, and cabbage for each experiment. She chopped the produce into uniform sizes and soaked it for 10-30 minutes in tap water, bleached solution, or deionized water as a control, in a salad spinner, spun the bowl every now and then to agitate the produce and the water.

She then began the work of analyzing the wash water: separating out the fruits and veggies, and concentrating the wash water, so that even low levels of DBPs could be measured. She analyzed the concentrated solution using GC-MS, a technique that separates mixtures of chemicals to measure the amount of each one individually.

At left, chopped strawberries; at right, strawberry wash water concentrated for analysis. Source: Wan-ning Lee, Georgia Institute of Technology


What Lee found is ultimately unsurprising: yes, having a chlorine wash step can significantly increase the formation of DBPs in the leftover produce wash water. This is especially troubling given that the wash water is often reused in produce processing facilities. Hopefully, with the improved methods developed in this paper, we’ll be able to track this problem across facilities more easily.

An important issue still remains, however: DBPs in the fruits and vegetables themselves. While this paper specifically focused on the leftover wash water, Lee also studied DBPs formed in the produce itself, with the results to be published in a future paper. Says Lee, “Now that we realize those carcinogenic chemicals can be formed during washing, it is important to know how much they accumulate in the produce. We must do both: control food-borne pathogens and reduce cancer risks from DBPs residues. The most important thing is to protect our food safety.”

Stay tuned to envirobites for an update on this story soon!

Share this:
Laura Mast

Laura Mast

I'm a PhD student in environmental engineering at Georgia Tech. Broadly, I study resource recovery: how we can look at waste products as untapped mines of valuable materials. Specifically, I develop methods to extract rare earth elements (required for everything cool/high tech/green we've made in the last 20 years, like LED screens, batteries, permanent magnets) from coal fly ash- a byproduct from burning coal for electricity. In my spare time, I run a start up called Populy, which provides electronic judging for STEM competitions, swing dance, and hang out with my dog.

Leave a Reply