Direct Heat and Superheated Steam Roasting System

[Hiroyuki]

As I briefly mentioned in another thread, Masashi Nakagawa of Flavor coffee in Aichi prefecture, Japan, was the first to succeed in roasting coffee beans by applying superheated steam. In what he calls the direct-heat and superheated steam roasting system, he applies superheated steam to coffee beans over direct heat in the drying phase (phase before the first crack) of the roasting process. He says that this allows him to control the humidity in the roaster, thereby controlling the moisture in the beans at will while keeping other components in.

This has been just an introduction to his unique system. In my next post, I think I'll provide some more information about it. I'd like to talk about a unique paper drip method called the Matsuya method later.

Masashi Nakagawa:

Flavor coffee:

Direct heat and superheated steam roasting system, with a super steamer on the right:

Edit to add Flavor coffee's website (Japanese only):

http://www.flavorcoffee.co.jp/index.html

Edited July 25, 2004 by Hiroyuki (log)

[phaelon56]

I'm really looking forward to hearing more. Fascinating concept and it seems to have great potential. Do you know of any English language sites that offer more detail on the overall concept and process?

[Hiroyuki]

He recalls why he has succeeded in completing his system while large companies have failed. (He says that there are a large number of companies working on the idea of applying superheated steam to coffee roasting.)

He was different right from the start.

Unfortunately, he was unable to produce superheated steam at high temperature, and he focused on increasing the humidity in the roaster at temperatures higher than the 'inversion temperature' (170 degrees centigrade).

On the other hand, large companies focused on the greatest property of superheated steam, high heat transfer efficiency, which means that they thought of using superheated steam as a heat source.

He attributes his success to devoting himself to controlling the moisture in the beans at will while keeping their other components in.

Some other information later.

[malachi]

i'd love to hear details on a comparative cupping (same beans, one roast with this method, one in a traditional commercial drum roaster).

[Hiroyuki]

Do you know of any English language sites that offer more detail on the overall concept and process?

I'm sorry I don't. I guess you mean the concept of the application of superheated steam to coffee roasting in particular, and I don't think there are sites that offer such information. Nakagawa says in his bulletin board that he doesn't know of any either.

He adds:

The existence of the inversion temperature seems to be well known.

Besides, it seems to be well known that due to its film condensation heat transfer, it (superheated steam) can offer a more efficient heating system than dry air.

Here I provide some links related to superheated steam in general, as well as some excerpts:

The following is from this site:

http://www.mec.ua.pt/_deptenglish/tema/unit/reslines7.html

Drying of Materials

The drying of materials with moist air and with superheated steam is also being worked. It was shown that the steam is more effective as drying agent above a critical temperature and that the moist air is the more effective drying agent below that temperature. This temperature is known as the vapour inversion temperature. Some research work is needed to well understand why the vapour is the most effective drying agent above the inversion temperature, as well as why the inversion temperatures reported in the literature are so different.

Another excerpt from this site:

http://www.osti.gov/energycitations/produc...osti_id=6103139

The rates of evaporation of water from a horizontal water surface into a turbulent stream of hot air or superheated steam at different free-stream mass fluxes and modulated temperatures were experimentally measured.^The pressure of the free stream was atmospheric.^For steam, the experimental results are mostly within 10% of the available analytical results.^Two previous experimental results about 50% and 300% higher than the analytical results.^For air, the measured evaporation rates consistently higher than the analytical results.^An estimate of the conduction heat transfer from the walls of the test section to water was made for several air tests.^If the conduction heat transfer were subtracted from the total heat transfer, the measured evaporation rates are actually quite close to the analytical results.^The present experiment also confirms the existence of a temperature, called the inversion temperature, below which the water evaporation rate is higher in air than in steam, but above which the opposite is true.^The inversion temperature is in good agreement with the analytical prediction.^The results for both air and superheated steam show that a certain scaled expression for the evaporation rate is independent of the free-steam mass flux, also in agreement with the analytical prediction.

An interesting passage from http://www.rsc.org/lap/educatio/eic/2002/clifford_mar02.htm

(Note, however, that this is a description of superheated water, not superheated steam.)

During any extraction, some of the components may undergo reactions, which may or may not be advantageous. In the extraction of explosives, for example, from decontaminated soil the explosives are degraded to benign substances.11 For plant materials at lower temperatures, oxidation and hydrolysis occur to a minor extent, which may be acceptable and even improve the product. The most interesting results are obtained with green Java coffee beans. Extraction with superheated water at ca 200°C produces a brown liquid, which has the aroma of coffee. After extraction the beans are whole and homogeneously dark brown: water appears to have permeated throughout the individual beans, they had expanded to about double their initial size. Analysis by head-space gas chromatography shows that the 'coffee' obtained has a much more concentrated extract, but the components and their proportions are very similar to those found in conventional coffee. Thus a process similar to roasting is occurring for the green beans. These experiments may give rise to a process for producing coffee flavours for the food industry.

Superheated steam drying technologies (pdf file):

http://www.ageng.ndsu.nodak.edu/asae/rrv/RRV03-0014.pdf

Another site on superheated steam (also a pdf file):

http://www.engr.usask.ca/societies/csae/c9914.pdf

[Hiroyuki]

i'd love to hear details on a comparative cupping (same beans, one roast with this method, one in a traditional commercial drum roaster).

Nakagawa says he doesn't employ the cupping technique.

He says that he uses the Matsuya paper drip method (which I'd like to describe later in another post) to examine what flavor he gets at a specific time.

The following is a summary of his reply to your post:

The greatest characteristic (of roasting with superheated steam) is that it produces strong (coffee beans).

The reason for this is, I think, that with ordinary roasting, the rate at which water evaporates from inside the bean differs from that at which it does from its surface, so that it is difficult to cause the same chemical change to the entire bean at once.

On the other hand, using superheated steam allows one to quickly get out moisture only. ... (omitted)

For this reason, the components of coffee are denser, I think.

The characteristic (of superheated steam) of getting water out is very effective to those coffee beans that are hard to get out moisture from, such as hard ones, but is not so effective to soft ones.

As you can see, anyone can roast soft beans...

Another thing I can say is that a large amount of hot air is required to heat coffee beans.

This means that other components (of coffee beans) are also removed when moisture is removed.

Superheated steam makes it possible to heat beans with a smaller amount of hot air.

And, it has a characteristic of not taking away other components when getting moisture out.

Thus, the components of green beans remain unremoved after the 'drying' phase.

...(omitted)

The reply continues, but I'd like to translate the remaining part after I explain the Matsuya method.

[Hiroyuki]

(Continued)

Another thing is that in the second half of the drip process with the Matsuya method, the various unpleasant tastes ( 'zatsu mi' in Japanese) are now less unpleasant.

Before I used this system, there was a clear distinction between the first half with flavorful tastes and the second half with various unpleasant tastes. ...(Omitted)

In short, it's drinkable.

This is probably due to the fact that the chemical change in the inside is similar to that on the surface.

The reply continues, but is not directly related to the topic.

***

On his website, he admits that his superheated steam technique is not yet completed. He says he has yet to fully understand how much steam to apply to get the best results.

He cites several potentials of his direct heat and superheated steam roasting system, although they are still under investigation, such as:

Elimination of robusta odor

Small-amount roasting

(He has succeeded in roasting 200 g of coffee in a roaster with a capacity of 3 kg.)

Elimination of smoke

[Hiroyuki]

Today, I have received a total of three e-mails from the manufacturer of the 'Super Streamer', in which they say they have no English-language literature on the product and that they have no plans to sell it overseas at the moment. They cite several reasons for their inability to sell it overseas, some technical and some marketing. I'm very sorry about that because I think it's a perfect product for those small coffee shops and cafes that wish to distinguish themselves from their competition, especially such giants as Starbucks.

Manufacturer's page describing the 'Super Steamer' (Japanese only)

http://www.maruyasu.co.jp/seihin/sangyo/kanetu/index.html

The 'Super Steamer' has been developed jointly by the manufacturer and Nakagawa of Flavor coffee.

[phaelon56]

He cites several potentials of his direct heat and superheated steam roasting system, although they are still under investigation, such as:

Elimination of robusta odor

Small-amount roasting

(He has succeeded in roasting 200 g of coffee in a roaster with a capacity of 3 kg.)

Elimination of smoke

I see enormous benefit for smaller shps in having the ability to roast smaller quantities. The short shelf life of roasted coffee and the minimum weight of beans required by many commercial roasters presents a problem for the smaller specialty shop. It's desirable to sell a good selection of varietal coffees but not if some goes to waste due to spoilage being able to roast a few pounds at a time of one coffee would enable a retailer to develope a trade of loyal cusotmers who come for their specific varietal or blend, which could be roasted in small quantities. The elimination of smmoke is a huge benefit for small roasters in urban areas - absolutely crucial.

I find the comment about robusta smell particularly intriguing. Hiroyuki - can you comment more on the prevalence of robusta use in Japanese coffee culture? Some large Italian roasters add a small percentage of good quality robusta to their espresso blends and handful of American roasters also do so (David Schomer's Caffe Vivace being one of the most well known US practicioners).

Robust has a bad reputation in the US market among discerning coffee drinkers for good reason becasue quality robusta rarely makes it to this market in any quantity. Some of the green bean purveyors will have Uganda Nanga farms robusta and on occasion some of the better Indian washed robustas (which are said to be very god). The reason I inquire about the Japanese market is because it is generally believed that that the vast majority of the annual premium robusta production goes to Japanese buyers, who are willing to pay a premium. Robusta has a higher caffiene content but even the better ones can have a bit of a bitter edge (not always a bad thing, especially in an espresso blend).

Are there specific blend types, desire for higher caffiene levels or other issues you're aware of which influence this or is it just tradition?

[coffeeroaster]

I see enormous benefit for smaller shps in having the ability to roast smaller quantities. The short shelf life of roasted coffee and the minimum weight of beans required by many commercial roasters presents a problem for the smaller specialty shop. It's desirable to sell a good selection of varietal coffees but not if some goes to waste due to spoilage being able to roast a few pounds at a time of one coffee would enable a retailer to develope a trade of loyal cusotmers who come for their specific varietal or blend, which could be roasted in small quantities.

I certainly agree, but super-small batch is not an impossibility now. I have built a small business roasting coffee 1 lb. at a time for very loyal customers who appreciate the freshness of coffee that isn't roasted until they order it. It does seem that "alternative" roasting technologies are more responsive to the needs/desires of those who want to roast in small quantities -- my machine is a commercial grade fluid bed (hot air) roaster, and while the method is often dismissed by drum "purists," it meets my needs better than the smallest commercial quality drum would, and my customers keep coming back for more.

My first reaction is that roasting with steam seems a contradiction, but far be it from me to poo-poo a new way of doing things. I will keep my eyes open for more info on this.

[malachi]

i don't quite see where the "small batch" problem lies.

there are excellent, commercial, drum roasters that can roast below 1lb dry charge weight (the Ambex Tabletop Roaster is a good example).

there are fluid bed roasters that roast 1lb dry charge.

there are sample roasters that go down even lower.

i currently roast in a 1K propane fired drum roaster and can roast down to .75lb dry charge weight with consistent quality.

[Hiroyuki]

can you comment more on the prevalence of robusta use in Japanese coffee culture?

Ｔｈｅ　Japanese are great consumers of canned, bottled, and instant coffees, and they are mostly made from robusta.

This is again a layman's comment.

[Hiroyuki]

My first reaction is that roasting with steam seems a contradiction,

I know how you feel. I took me a whole week or so to fully understand the properties of superheated steam and appreciate the true value of its application to coffee roasting. The key word is the inversion point, above which superheated steam causes more water to evaporate than dry air does.

***

Let me show you a brochure (pdf file) on the Super Steamer although it is entirely in Japanese.

http://www5f.biglobe.ne.jp/~nanacafe/image/maruyasu0.pdf

At the top of the second page, you can see two photos. The right one shows a bean roasted with superheated steam, and the left one a bean roasted without it.

Below the photos, you can see a 3-axis graph. The vertical axis represents flavor (umami), the right one bitterness, and the left one acidity.

The blue points indicate the results of beans roasted with superheated steam, and the pink ones the results of beans roasted without it. You can see that superheated steam increases both umami and bitterness.

Edited August 6, 2004 by Hiroyuki (log)

[phaelon56]

I find it most intriguing to see mention of a new "microwave" oven that uses superheated steam and convection heat to cook.

New Microwave thread

The sellign point is supposed to be that it cooks the fat out of the food but I'd have to wonder if it could be adapted to home roasting of beans using the superheated steam and convection as well as condensation heat (it's not clear to me whether it also uses ordinary microwave rays as well). Obviously venting of smoke would be an issue.

[Hiroyuki]

Really intriguing. It's a good example of the application of superheated steam as a heat source, not as an auxiliary measure as in the direct heat and superheated steam roasting system developed by Nakagawa.

I'm not 100% sure, but from what I read from Sharp's website, this appliance seems not to use the microwaves. Anyway, to satisfy your curiosity, I have just sent an inquiry to the manufacturer, asking if they have an English literature on this product.

Sharp's webpage on the product (Japanese only):

http://www.sharp.co.jp/healsio/index.html

[Hiroyuki]

I have just received a reply from Sharp. They say that the appliance cannot be used to roast coffee beans because it does not have an agitator.

They provided a link to their English webpage on this product:

http://sharp-world.com/corporate/news/040823.html

[Hiroyuki]

Part of Nakagawa's comments on this home appliance:

I have seen several superheated stream ovens before (all commercial ones).

When I tried to roast coffee beans in a superheated stream oven, this didn't work well.

When I attempted to roast them for 14 minutes or longer as with regular roasting, moisture was removed completely and the beans didn't crack.

When I attempted to roast them for three minutes, they did crack, but they didn't have aroma. And the flavor was not very good.

[Hiroyuki]

The rest of Nakagawa's comments:

Unfortunately, it was on that occasion only that I made roasting experiments using a superheated steam oven, so I felt like trying some more…

If you are to roast coffee with superheated steam only, you have to shut off the exhaust completely and return (the oven to the initial state?) to re-heat the steam; otherwise, you can’t get a sufficient quantity of heat.

That’s why I focused on the inversion point of superheated steam only.

(I think that Sharp focused on film condensation heat transfer and defatting in the absence of oxygen.)

(Responding to another person’s post)

As for the difference in taste, which Yankee mama mentioned…

Quite honestly, using superheated steam causes changes in taste, which some people like while others don’t.

This is true.

The reason for this is that using superheated steam causes the taste to be by far stronger than regular roasting.

So, those who like sappari (bland) taste tend to dislike it.

(I, for one, think it is better to cause thorough chemical change…)

What is interesting is that the aroma decreases both when the time to apply superheated steam is too long and when it is too short.

So, at this point, you can never be sure how much steam to supply to get the best results unless you make numerous experiments.

In my roasting, I supply steam for six minutes.

I think that the amount of steam should be adjusted according to the type of bean…

At this point, however, I have not yet been able to experiment up to that level.

(I expect the time to fall within five to seven minutes.)

This has been my general comments on superheated steam.

***

(I must confess that I really cannot translate his style of writing to equivalent English.)

[Hiroyuki]

As for the difference in taste, which Yankee mama mentioned…

Yankee mama seems to be an old friend of Nakagawa’s. Here is what she wrote on Nakagawa’s bulletin board recently:

Now that Nontan’s (= Nakagawa’s) coffee is stronger, I make it weaker before I drink it. So, once you brew it, you can keep it longer.

It’s economical.

--

I wrote jokingly to Yankee mama, “I can post your photo if you like,” and the result is…

a self-portrait?

Edit to add:

Sorry for getting off topic...

Edited August 31, 2004 by Hiroyuki (log)

[Hiroyuki]

An important message for phaelon56 and torakris from Yankee mama

Ever since I started posting on Nakagawa’s bulletin board, Yankee mama has often responded to my posts, sometimes referring to you, phaelon56 and torakris.

To make a long story short, Yankee mama has this important message for you, (which was originated half jokingly in Japanese by me, approved for posting on eGullet by Yankee mama, and translated into English by me):

***

1. Come to Japan and drink Nontan’s (= Nakagawa’s) coffee!! I love you!!

2. If you can’t, let me send some coffee beans from Japan. Just drink!!

***

This is not a joke, and I'd like both of you to take her offer seriously.

My suggestions would be to post your preferences (bean type, roast level, whether to grind or not, etc.) here and give your addresses to Flavor coffee at

master@flavorcoffee.co.jp

And I'd like you to post your candid opinions here.

IMPORTANT: I am not affiliated with Flavor coffee, except that I am fascinated by Nakagawa's innovative idea about roasting, his superb brewing techniques, and above all, his commitment to hand-picking all deficient beans after roasting.

[Hiroyuki]

his commitment to hand-picking all deficient beans after roasting.

Let me elaborate a little bit on this. At Flavor coffee, green beans are circulated through an electronic sorter to remove defects. (One pass through the sorter is not enough. Nakagawa developed a system that enables green beans to circulate through the sorter.) After roasted, the beans are circulated through the sorter again. Then, all defects are hand-picked, as shown in the photo below.

Edited September 8, 2004 by Hiroyuki (log)

[phaelon56]

***

1.  Come to Japan and drink Nontan’s (= Nakagawa’s) coffee!!  I love you!!

2.  If you can’t, let me send some coffee beans from Japan.  Just drink!!

***

My suggestions would be to post your preferences (bean type, roast level, whether to grind or not, etc.) here and give your addresses to Flavor coffee at

master@flavorcoffee.co.jp

And I'd like you to post your candid opinions here.

I sincerely wish that I could visit Japan but it may be about two years before such an opportunity presents itself. I'd be delighted to try this coffee and offer my candid opinions. At home I use whole bean coffee only which I grind when needed by the pot or if making espresso I grind by the espresso shot. My general preference for a roast level is for something just a bit lighter what is generally known as Full City. This would be medium dark with perhaps a few drops of oil visible here and there on the surface a few days after roasting but not really a dark roast.

I will forward you the other appropriate information by PM and also to the address suggested for yankeemama.

[Hiroyuki]

phaelon56:

Thank you for your reply. I need to ask a favor of you. Nakagawa says he doesn't understand English. (He says he immediately deletes any email he thinks irrelevant.) So, could you send the other necessary information including your address to me by PM? I'll translate all the necessary information and send it to Nakagawa.

torakris:

Have you decided whether to accept Yankee mama's offer? I really hope you accept it. Like I told phaelon56, Nakagawa doesn't understand English, so could you send all the necessary information including your address to me by PM?

To you both:

I promise to keep your personal information strictly confidential and delete it from my PC as soon as the offer is completed.

Thank you in advance.

[melkor]

The beans showed up yesterday afternoon, and this morning we pulled shots with the espresso blend and with the mocha beans.

These are both the espresso blend. I suspect the espresso blend is made mostly with columbian beans, it's a nice cup, extremely mild (needs no sugar or milk), and well balanced.

The Mocha as espresso required a touch of sugar, it has nice toffee and chocolate notes, I think it will be much better as brewed coffee.

More to come as I get to the others.

[phaelon56]

Just to clarify the context of Melkor's post (mea culpa - I should have done this sooner but had no idea he'd get the beans so quickly).... eGulleteer Hiroyuki was kind enough to send coffee samples that Mr. Nakagawa offered. I split up and distributed them by Priority Mail among some Coffee forum regulars who might serve as good judges of the beans roasted by this unique method. Apologies for not making it a public offer but there was a limited amount of coffee to distribute and it was crucial that I get the samples to people in a timely manner.

The beans samples were roasted exactly one week ago today if my understanding is correct.

I have also sampled the espresso and agree with your findings. It was smooth with very little bitterness and decent but not outstanding crema. I find it to be well balanced - no strong fruit or chocolate notes. More notes later - my schedule has been to busy to make coffee at hoem for the past few days.