Question from September's Chemical Processing
A bulk spray dryer operates between 2,800–3,200 psig with a bioproduct feed slurry (Figure 1). The dried solid auto-ignites at 430°C, and this has occurred twice in four years. The reliability is poor. The on-stream factor is about 75% for numerous reasons: erosion of the single spray nozzle, which lasts only about three days before pressure drops below 2,800 psig; build-up of solids — knockers were deactivated due to poor maintenance; shaker failure in the baghouse; repeated bearing failure because of dust and poor location of blowers; frequent gas leaks, especially during start-ups; and failure of the oil seals on the high pressure pumps. In addition, we have observed several safety issues, including operating without an inlet filter element, removing the cartridge filters on the pump to achieve a higher atomization pressure, and infrequent tune-ups of the burner. Any suggestions on how we can improve safety and reduce breakdowns?
Figure 1. Spray dryer suffers from poor reliability and poses operating dangers.
Replace the spray dryer
It almost sounds like this application isn’t well suited for a spray dryer. For harsh applications such as you have described, I would suggest looking into a rotary tray “turbo” dryer. Slurry is added to the top tray of the dryer. After one turn, the material is squeegeed to the next tray below. After several revolutions the product is dry. A unique aspect of the dryer is that material is agitated as it cascades between levels, exposing fresh surfaces for drying. Product is removed at the bottom by a screw conveyor or other means. The air flow is generally counter-flow, not co-current as in the spray dryer. Dry hot air enters the bottom and humid air exits the top. Very little dust accumulates. A baghouse isn’t needed. Another advantage is that turbo dryers are continuous; the spray dryer described is a batch dryer.
In one turbo dryer application, a urea plant processed 5,000 lb./hr. with an initial moisture of 20% to a final content of 0.2% through a dryer 47-ft. high with a diameter of 15-ft. In another example, 4,200 lb./hr. of catalyst pellets were dried from 45% to 18% in a 23-ft. dryer with a diameter of 19-ft. Tray dryers require more space than spray dryers but excel in efficiency, reliability and safety. A turbo dryer can dry any product below the boiling/flash point without the need for high pressure pumps. Installations I have been involved in have operated for years without any down time.
Scott E. Crosby, president
SKL Process Solutions, Grass Lake, Mich.
Change the nozzles
Have you ever considered air-atomizing nozzles? Significantly less line pressure is needed to form a spray pattern because the atomization energy is provided by compressed air or nitrogen.
Mark Otterson, process engineer
Lubrizol, Avon Lake, Ohio
Consider four steps
Surely there are many aspects to be corrected:
- change to multiple spray;
- use polytetrafluoroethylene nozzles with a pattern that will allow working at lower pump pressures;
- change the oil gaskets on pumps, and bearings materials; and
- add to the slurry, if possible, a rinsing aid that will allow better wet and dry flow.
Emilio Malaguti, technical manager
Chemtron, Hialeah, Fla.
Go to vacuum operation
Why not use a vacuum dryer? Do the spraying under vacuum and at much lower temperatures. Even a modest vacuum can lower the vapor pressure of the solvent significantly. For example, water boils at 212°F at 14.7 psia. With a partial vacuum of about 2 psi., about 100 torr., it boils at 205°F. A more substantial vacuum may help reduce the chance of a fire.
It’s tough to discuss further options without knowing more about the process and product and specific requirements. What's with this 2,800-psig. operating pressure?
Rick Vogel, senior process engineer
Fluor Enterprises, Greenville, S.C.
Explore problems in the lab
I've spent much of my 35 years in industry working with spray drying. To ensure the safety of a spray drying operation, lab data needs to determine the temperature of the exotherm. This is normally done in a Grewer oven (but also can include DSC/TGA data), which slowly ramps the temperature and monitors for exotherms. This will determine the safe operating temperature. In my company what is considered safe is a maximum operating temperature 50oC maximum below the exotherm temperature.
This can be reduced to 20oC with a modified test where a thicker bed of material is
used. Once the safe temperature is determined, the interlocks should be set up to insure that temperature is not reached. In the case of the fires, the temperature recordings should then examine the temperature recordings to see if they can assign a cause to the fires. My experience is that many companies, particularly smaller companies, have little expertise in this area and often operate in unsafe temperature regions.
On the issue of considering alternative dryers, there has to be a great deal of thinking in regard to the right dryer. First, a set of lab tests need to be run in the lab to determine the drying characteristics of the material. These include the safe drying temperature tests, drying kinetics in several different possible tests, stickiness of the powder at various moisture levels, caking behavior, granulating behavior, and sorbtive equilibrium tests. Once the characteristics are determined, the boundary conditions are needed ─ volume of production, location, etc. However, the most important property is the type of product. For example, do you need one that is granulated or individual particles? Is your goal a particular particle size? What kind of flow characteristics are desirable? Free flowing behavior is best. What are the dissolution requirements? What is the maximum moisture content, etc.? Using the drying characteristics and the boundary conditions together with expert knowledge of the various dryers, an appropriate dryer or dryers can be selected as candidates with a high likelihood of success. I've seen a lot of money wasted developing drying processes which have no chance to succeed because either the drying characteristics or the end use requirements were not considered carefully enough.
Dr. Rudy Lisa, sr. research associate
BASF Corp., Mt. Olive, NJ
Form a game plan
Clearly, this dryer is an accident waiting to happen. Consider these ideas to improve the safety and the reliability of the dryer:
- a new spray nozzle material;
- reinstall the existing knockers and add others later after inspection;
- create a temperature monitoring alarm system;
- monitor blower bearings and consider alternatives;
- install gas sniffers at top of dryer;
- consider replacing pneumatic burner control valve with electric actuator;
- evaluate alternatives for eliminating unreliable positive displacement pumps;
- evaluate pressure losses in system to improve blower performance;
- consider elimination of unreliable shaker arms in baghouse;
- inspect fire control systems; and
- conduct time study on shut-downs and clean-outs.
Apparently, the spray nozzle material is eroded by the slurry. This isn’t surprising. In drying a slurry, the droplets must be very fine. To quote Williams-Gardner’s Industrial Drying (Leonard Hill, London, 1971), p. 56: "A pump-able slurry, paste, or sludge will require the maximum possible immediate moisture evaporation on its introduction into the dryer in order that it may break down into manageable particles of semi-dried material and not build up on the internal surfaces of the dryer." Without sufficient stress, the slurry will cake the inside of the dryer, becoming a fire hazard as fine dust break free from the cake and forms into tinder material. I would recommend using nickel or a nickel-alloy for the spray nozzles.
Replace or repair the existing bin activators or knockers. Add knockers to the lower-middle cone of the dryer itself. Don’t add knockers near the base of the dryer — this could promote flaking in the cake at the bottom of the dryer. This cake will begin to form regardless of how clean the operation is. There’s no point in kicking a sleeping dog!
Initiate a "check-weight" program. Every time the dryer is shut down to collect the product, weigh the material collected where it’s not supposed to be: the duct, the bottom of the baghouse, cyclones, and dryer bottom. You’re making progress when the bag weight decreases. Add additional knockers as needed based on this study.
Operations is flying blind. The most likely spot for a fire — the bottom of the dryer — is neglected in the present layout. Add two thermocouples 180° apart at the bottom of the cone above the 180° elbow. Add a thermocouple near the top of the dryer to detect fluctuations in the burner. Add another thermocouple downstream of the 180° elbow to detect a rise in heat as the product catches fire in the bottom of the dryer: heat follows flow.
Gas in a confined space is a recognized danger. Dryers tend to be installed in ceiling areas where dry air removes moisture from product leaking from high-pressure line. This is a serious potential hazard. Install multiple gas sniffers, check the sprinkler system and look at pipe expansion at the burner inlet: a common source of leaks.
Electric actuators can outperform pneumatics in fine controls. This is worth looking into for the burner temperature control valve.
Perhaps there’s an alternative to high pressure. Maybe a portion of the combustion air could be used to atomize the slurry. Gas/liquid atomizers often operate at a lower pressure than high pressure liquid. This idea only works if high pressure compressors are not involved, otherwise you risk poor reliability.
Building up product in ducts is mainly a problem to flow. Backpressure will cause a centrifugal compressor to back down its curve, producing less fluidizing gas: solids fall out of suspension. Investigating the entire system for pressure losses and dead areas will eliminate this potential problem.
Shaker arms are often replaced by more reliable pulse-jet systems. This may not work here because the blower capacity may be limited but it’s worth investigating.
If it hasn’t been done recently, the fire control system should be inspected.
Clean-out is inevitable with this type of batch dryer. There should be ways to reduce downtime. I would do several detailed time-studies on this operation to identify ways to make the operator’s job easier and thereby increase the run time for the dryer.Dirk Willard, senior process engineer
Ambitech Engineering, Hammond, Ind.
A hydrolysis catalyst bed is designed for removing COS, H2S and HCN at 430°F from 40,000 lb./hr. of H2 gas at 400 psig. The diameter of the packed bed is 5 ft. The maximum allowable heat-up rate for the alumina catalyst is 90°F/hour. At the extreme minimum ambient temperature, 20°F, it will take almost five hours to heat the bed safely. One idea proposed for heating the bed is to recirculate gas through a mineral oil heater and the bed until the catalyst can safely reach reaction temperature. An estimate of the heat-up time for the catalyst was made using Schumann’s curve and the method described on pp. 668–670 of Kern’s Process Heat Transfer. Unfortunately, an estimate of the rise in the bed by recirculating the gas indicates that it will take only 53 minutes, which is a gradient of 464°F/hour. Besides this problem, the largest practical blower that could circulate low-pressure gas would need over 1,000 hp. and could only operate up to 15,000 lb./hr. at 3 psig. Further work with the Schumann curve showed that it could take over 16 hours to achieve temperature with such a blower. After discussion, another start-up problem emerged: condensation of water on the catalyst. What can you recommend to ameliorate the condensation problem? Because a blower is not practical, what other ideas can you suggest?
Send us your comments, suggestions or solutions for this question by December 10, 2007. We’ll include as many of them as possible in the January 2008 issue and all on CP.com. Send visuals — a sketch is fine. E-mail us at ProcessPuzzler@putman.net or mail to ProcessPuzzler, Chemical Processing, 555 W. Pierce Rd., Suite 301, Itasca, IL 60143. Fax: (630) 467-1120. Please include your name, title, location and company affiliation in the response.
And, of course, if you have a process problem you’d like to pose to our readers, send it along and we’ll be pleased to consider it for publication.