A work-in-progress while I am figuring out the best way to make RSSCT methods useful to biochar water treatment practitioners out there in the real world…
Sorry for the radio-silence over the past couple of weeks. I hit a hurdle on the next chapter section and it’s taking me some time to work my way over it.
You probably have noticed that previous sections on rapid small-scale column tests (RSSCTs) have mentioned the two predominant design approaches - constant diffusivity (CD) and proportional diffusivity (PD). Again, this refers to the assumption we make about whether the diffusion coefficient of target pollutant molecules changes when adsorbent granules are crushed from the ~1 mm diameter particles used in full-sized treatment systems and pilot columns to the ~0.1 mm diameter particles used in RSSCTs.
I’ve repeatedly alluded to CD and PD approaches to the RSSCT without specifying which one is “correct” or “the best.” The reason is that, despite going on four decades of research, we still don’t have a one-size-fits-all approach to obtaining accurate simulations of real-world treatment systems using RSSCTs. The culprit - I bet you can guess - is fouling by dissolved organic matter (DOM). DOM fouling is complex, multi-modal, varies over time during the service life of an adsorbed unit, and is spatially heterogenous with depth in the adsorbent bed. The diffusional models of target pollutant mass transport within adsorbent particles can’t capture the complicated and multiplicative effects of DOM fouling very well.
Accordingly, pollutant breakthrough profiles obtained from RSSCT studies typically do not overlay pollutant breakthrough profiles from pilot columns or full-sized treatment units. RSSCTs over- or under- estimate (usually overestimate) treatment capacity at full-scale, and vary in their ability to capture adsorption kinetics - which is to say, the shapes of RSSCT breakthrough profiles differ (are steeper or more shallow) than those obtained from pilot and full-sized units.
Over the years a bevy of methods for scaling RSSCT data - i.e., adjusting (a cynic might say “fudging”) RSSCT breakthrough profiles to better match pilot/full-scale - have been elaborated. These include data fitting and transformation approaches using diffusional models such as the pore and surface diffusion model and the homogeneous surface diffusion model; methods for estimating “DOM fouling factors” from target pollutant physical-chemical properties and background water chemistry parameters; and various combinations of computational and empirical approaches, for example using equilibrium data from batch tests and/or diffusion coefficients obtained from kinetic tests, etc.
And like the question of “CD or PD?” all these different scaling approaches have advantages and disadvantages, work in some cases and not in others, and haven’t, at least yet, led to a clear “one best approach for all (or most) circumstances.”
All this is a lot to mull over, and I’ve been mulling it over for years.
This was the hurdle I hit after writing the last section. I arrived at a conceptual mind-block: How to structure a discussion of biochar RSSCT studies that practitioners will find useful, user-friendly, applicable, and reasonably concise and to-the-point.
I think I’m just now figuring out how to do this, in a way that makes the most sense for the book.
I’m putting together a summary of pluses and minuses, strengths and shortcomings of the most common RSSCT design and scaling approaches as decision assistance for readers who want to conduct your own RSSCT studies.
Along with this I’m coming up with a guide to a recommended scaling approach for using RSSCT data (CD or PD) to make conservative projections for how full-scale treatment systems would likely perform - for readers who want to use existing RSSCT datasets to develop treatment system design scenarios and options.
That’s what’s coming to the subscribers’ list over the next couple of weeks. Thanks for your patience with the delay.
One reason it’s taken a while is that en route I’ve needed to build a database from published studies that collected both RSSCT (CD or PD, or both) data as well as data from pilot columns or full-sized systems treating similar water and pollutants. For biochar, there aren’t very many studies out there, so I have turned to the granular activated carbon (GAC) literature to compile a larger and more robust dataset from which to draw conclusions. Surprisingly, perhaps, there isn’t a one-stop-shop for these datasets (though one or two other researchers have done excellent and very helpful work building partial datasets) so I’ve been compiling a master database myself. The image at the head of this entry alludes to this work-in-progress, and will be elaborated in due time.
And though work delays can be frustrating, I have at least one happy reason for having been pulled away from the number crunching for a few days over the last week or so. Here at the farm one of our sows had piglets! About ten days ago, discovered by my wife during evening chores and feeding.
<1 day old kune kune piglets.
This is the second batch of surprise piglets to arrive this summer. Back in late April we got a few full-grown sows from a nearby farmer who was just about giving them away. Turned out they were pregnant. What started out as a small project, “Let’s raise a couple of pastured hogs!” turned into “Wow, now we have 10 pigs!” We have been hard at work getting a new shelter and expanded pasture set up for these new members of our little Appalachian hilltop menagerie.
Mama and li’l piggos enjoying the new digs.