第51卷第5期化工學(xué)報(bào)Vol. 51 No5000年10月Journal of Chemical Industry and Engineering( ChinaOctober 2000研究論文變壓吸附空分制氮過(guò)程的研究盧洪李成岳北京化工大學(xué)化學(xué)工程學(xué)院,北京100029)摘要建立了一套中試裝置,對(duì)以商業(yè)炭分子篩為吸附劑的變壓吸ⅨA)空分制氮循環(huán)過(guò)程進(jìn)行了系統(tǒng)研究.用所建立的數(shù)學(xué)模型對(duì)相應(yīng)實(shí)驗(yàn)進(jìn)行模擬并將模型預(yù)測(cè)與實(shí)驗(yàn)數(shù)據(jù)進(jìn)行比較,結(jié)果表明模型是可靠的關(guān)鍵詞空氣分離變壓吸附氮?dú)庵袌D分類號(hào)TQ28.5引言6對(duì)變壓吸附循環(huán)過(guò)程, Raghavan12Hassan3H及Faro5均進(jìn)行過(guò)模型預(yù)測(cè)與實(shí)驗(yàn)的對(duì)照研究,但所涉及裝置均為小型的實(shí)驗(yàn)裝置,其吸附柱直徑僅為幾cm,吸附柱內(nèi)工況與工業(yè)裝置存在較大差異.另外在相關(guān)文獻(xiàn)中也未見(jiàn)到對(duì)吸附柱5內(nèi)氣相濃度分布進(jìn)行實(shí)驗(yàn)測(cè)定并與模擬結(jié)果進(jìn)行比較的報(bào)道.顯然,僅僅將實(shí)驗(yàn)中吸附柱出口的濃度與模型預(yù)測(cè)值進(jìn)行比較去證明模型的可靠性是不夠Fig. I Flow sheet of PSA unit for nitrogen production的.本文報(bào)道在中試裝置上,在接近工業(yè)裝置的操-compressor : 2--water cooler 3-condlensate collector作條件下,測(cè)定吸附塔軸向溫、濃度分布,并對(duì)整4-drving towers 5-12-solenoid valves 13-buffer tank個(gè)循環(huán)過(guò)程的特性進(jìn)行模擬與實(shí)驗(yàn)研究的結(jié)果14--flowmeter 15--pressure regulating valve 16-adsorptioncolumns 17-20-check valves 21-28-solenoid valves1實(shí)驗(yàn)29--nitrogen tank實(shí)驗(yàn)裝置的工藝流程如圖1所示,空氣預(yù)處理調(diào)壓閥調(diào)節(jié)至指定壓力并經(jīng)轉(zhuǎn)子流量計(jì)計(jì)量后作為包括壓縮及脫水,原料空氣經(jīng)壓縮機(jī)升壓后進(jìn)入間產(chǎn)品氣體引出;另一路計(jì)量后進(jìn)入再生塔進(jìn)行吹掃壁式水冷器使壓縮空氣中的水部分冷凝,然后進(jìn)入再生,反吹計(jì)量管路由4個(gè)單向閥和1個(gè)轉(zhuǎn)子流量干燥器進(jìn)一步吸附脫水.干燥部分由兩個(gè)交替使用計(jì)組成的硅膠干燥器組成,當(dāng)其中一個(gè)處于吸附狀態(tài)時(shí)實(shí)驗(yàn)裝置的循環(huán)方式與大多數(shù)工業(yè)裝置的實(shí)際從該塔引岀部分干燥空氣對(duì)另一干燥塔進(jìn)行吹掃再操作相冋,由4個(gè)階段組成:加壓吸附→均壓(降生,反之亦然壓)放空吹掃→均壓升壓)加壓吸附干燥空氣經(jīng)減壓閥減壓后進(jìn)入緩沖罐,經(jīng)過(guò)轉(zhuǎn)1.1吸附床及吸附劑規(guī)格子流量計(jì)計(jì)量后,由調(diào)壓閥調(diào)節(jié)至指定壓力進(jìn)入吸實(shí)驗(yàn)裝置吸附床尺寸及吸附劑的有關(guān)數(shù)據(jù)如附塔進(jìn)行吸附分離.氣體的流動(dòng)均由電磁閥控制,下:裝置額定產(chǎn)氣量1.0mh-1,每床填充德國(guó)產(chǎn)其中電磁閥23、24用于均壓操作,離開(kāi)吸附塔的BF型炭分子篩吸附劑7.5kg,床層空隙率0.32,吸氣體分為兩路,一路經(jīng)出口閥進(jìn)入氮?dú)赓A罐,由附塔高800m,吸附劑有效填充高度720mm,塔徑159mm;吸附劑顆粒尺寸為φ2mm×2~5mm199904-09收到初稿,199908-24收到修改稿顆粒表觀密度為0.84×10kgm-3.由于吸附塔直聯(lián)系人:李成岳.第一作者:盧洪,男,27歲,碩士,工程徑比一般實(shí)驗(yàn)研究所用吸附柱要大得多,可以認(rèn)為第51卷第5期盧洪等:變壓吸附空分制氮過(guò)程的研究1.2溫度測(cè)量層溫度波動(dòng)很小,對(duì)吸附分離幾乎不產(chǎn)生影響,無(wú)為了考察吸附塔內(nèi)軸向溫度分布及熱效應(yīng)的大論操作條件如何改變,軸向溫度分布曲線基本在環(huán)小,在吸附塔的塔壁上沿軸向不同位置開(kāi)設(shè)7個(gè)測(cè)境溫度附近浮動(dòng),上下不超過(guò)3℃,如圖2(a溫口,分別安裝E型鎧裝熱電偶,其測(cè)量端沿徑圖xb所示向伸至吸附塔軸線位置.熱電偶進(jìn)行冷端補(bǔ)償后直表觀吸附熱效應(yīng)與基礎(chǔ)實(shí)驗(yàn)中所測(cè)得吸附熱較接與PCL-818L及PCID-789D數(shù)據(jù)采集卡連接,小的結(jié)果是相容的,因此,吸附熱效應(yīng)在建立變壓用一臺(tái)微機(jī)對(duì)吸附塔溫度進(jìn)行實(shí)時(shí)跟蹤測(cè)量并吸附空分制氮模型時(shí)可以不予考慮6]記錄2.2數(shù)學(xué)模型1.3濃度測(cè)量由于本系統(tǒng)的熱效應(yīng)較小,因此,在建立數(shù)學(xué)為了測(cè)定吸附塔內(nèi)氣相氧濃度的軸向分布,沿模型時(shí)可不考慮熱量衡算,而僅僅考慮吸附塔內(nèi)的吸附塔璧在軸冋不同位置處開(kāi)了7個(gè)采樣口,待體質(zhì)量衡算,本文所建立的吸附塔數(shù)學(xué)模型見(jiàn)附錄系達(dá)到循環(huán)定態(tài)后,打開(kāi)指定位置的采樣閥,與便利用所建立的數(shù)學(xué)模型,對(duì)實(shí)驗(yàn)裝置的性能進(jìn)攜式CY-丌B型數(shù)字測(cè)氧儀連接,對(duì)同一周期內(nèi)幾行了模擬與預(yù)測(cè).涉及到的一些操作參數(shù)及模型參個(gè)特定時(shí)刻的濃度值進(jìn)行測(cè)定,直到數(shù)據(jù)穩(wěn)定,然數(shù),確定方法如下后對(duì)下一測(cè)點(diǎn)相應(yīng)時(shí)刻的濃度值進(jìn)行測(cè)定,直到取23吸附塔壓力得所有測(cè)點(diǎn)在對(duì)應(yīng)時(shí)刻的氧濃度數(shù)據(jù)為止.根據(jù)這為了確定各循環(huán)階段壓力變化情況,實(shí)驗(yàn)過(guò)程些數(shù)據(jù)即可繪岀吸附塔內(nèi)氣相氧濃度隨時(shí)間的變化中,利用壓力表對(duì)各操作階段吸附塔內(nèi)壓力隨時(shí)間情況及軸向分布曲線.變化情況進(jìn)行了跟蹤測(cè)量,得到各階段壓力變化曲1.4流量測(cè)量線.對(duì)壓力曲線進(jìn)行擬合,即可得到壓力變化關(guān)聯(lián)根據(jù)實(shí)驗(yàn)研究的需要,需對(duì)3處的流量及壓力式,用于模擬計(jì)算進(jìn)行測(cè)量,以確定系統(tǒng)的原料氣消耗量、吹掃氣用2.4吸附平衡常數(shù)及擴(kuò)散時(shí)間常數(shù)量及產(chǎn)氣量,從而確定本系統(tǒng)的性能指標(biāo).如圖1對(duì)N2、O2在炭分子篩(CMs)上的吸附基礎(chǔ)數(shù)所示,所有流量計(jì)量均使用轉(zhuǎn)子流量計(jì).由于循環(huán)據(jù),采用文獻(xiàn)6沖所述實(shí)驗(yàn)結(jié)果,如表1所示操作,進(jìn)料及吹掃流量均會(huì)隨著操作階段旳轉(zhuǎn)換而Table 1 Equilibrium and diffusion time出現(xiàn)較大波動(dòng),因此在實(shí)驗(yàn)數(shù)據(jù)的處理過(guò)程中只能constants of N2 and O,( 26C)近似地采用平均流量2實(shí)驗(yàn)結(jié)果及與數(shù)值模擬的比較177×10-31.17×10-42.1吸附床層的熱效應(yīng)2.5線性推動(dòng)力傳質(zhì)LDF模型中的Ω值在實(shí)驗(yàn)過(guò)程中,對(duì)床層溫度分布進(jìn)行跟蹤考察模擬過(guò)程中,采用了實(shí)驗(yàn)確定Ω值的方法表明,隨著吸附或解吸的進(jìn)行,床層內(nèi)溫度分布的以表2中實(shí)驗(yàn)10所列實(shí)驗(yàn)數(shù)據(jù)來(lái)標(biāo)定aA及B的變化趨勢(shì)合理反映岀吸附過(guò)程放熱和解吸過(guò)程吸熱值,然后用于模擬計(jì)算的特性.但從總體上來(lái)看,在吸附及解吸過(guò)程中床循環(huán)過(guò)程特性數(shù)值模擬與實(shí)驗(yàn)數(shù)據(jù)的比較見(jiàn)表303.03020302.0301.0301.03000299.0299.000.20406081.000204060.81.0(a)adsorption step(b) purge steFig 2 Bed temperature588化報(bào)2000年10月Table 2 Comparison between experimental and simulation resultsxA/%( ProductRA2l,130.120.0010.1860,1880.15602.0.300.350.226.786003.560.330.371.680.3682.1601.67400.290.250.2440.251,70380.290.2630.2643.150.370.2840.2由表2可以看出,無(wú)論是產(chǎn)品純度還是氮?dú)饣厥章?模擬計(jì)算都能較好地預(yù)測(cè)實(shí)驗(yàn)結(jié)果,幾乎所有模擬計(jì)算均能正確反映操作條件變化對(duì)性能的15影響下面進(jìn)一步說(shuō)明模型預(yù)測(cè)結(jié)果的可靠性圖3表明吸附塔岀口氧含量隨操作周期數(shù)增加而不斷變化的情況.隨著循環(huán)操作的進(jìn)行,吸附塔10203040出囗處的氧含量從21%很快下降,大約經(jīng)過(guò)25個(gè)循環(huán)后,出口處濃度不再改變,從而達(dá)到循環(huán)定Fig 3 O2 molecule fraction at outlet is cycle numbe態(tài),模擬結(jié)果與實(shí)驗(yàn)數(shù)據(jù)吻合(t0a=3.55cms-,toe=1.67cms1,p=3.1×10°Pa同一周期中床層內(nèi)濃度分布濃度波的移動(dòng)情Pp=1.0×103Pa,tn=60s,tn=2s)況洳圖4所示,在加壓吸附階段,入口端漸趨飽和,吸附區(qū)域逐漸向出口端移動(dòng);在吹掃階段,在圖5及圖6分別表示體系性能隨流速比αN2的置換作用下得到清洗,吸附劑床層得以解吸(viv)和保留時(shí)間x(Bv4)的變化而改變,再生,無(wú)論是吸附階段還是吹掃階段,數(shù)值模擬結(jié)結(jié)果表明α和τ增大均有利于產(chǎn)品純度的提高,但果與實(shí)驗(yàn)數(shù)據(jù)都很好吻合,因此,模型預(yù)測(cè)較好地氮?dú)饣厥章蕜t相應(yīng)降低反映了實(shí)際操作中的這一動(dòng)態(tài)過(guò)程.圖7表明在特定的流速比a和保留時(shí)間r條件35251002040.60.81000.2040.6081.0(a)O2 molecule fraction profiles of adsorption step(b)O2 molecule fraction profiles of purge step(1oa=3.55cms1,to4=1.67cms1,t=60s,t=2s,p=3.1×10Pp=1.0×105Pa)第51卷第5期盧洪等:變壓吸附空分制氮過(guò)程的研究589有指導(dǎo)作用0.35(2)在BF-CMs吸附劑上,熱效應(yīng)對(duì)PSA空030分制氮過(guò)程操作性能的影響可以忽略(3)線性推動(dòng)力傳質(zhì)模型LDF,帶Ω因子)以很好地用于預(yù)測(cè)BF-CMS上PA空分制氮過(guò)程0405060.70.80.91.0的操作特性(4)在給定產(chǎn)品純度及其他操作條件后,適當(dāng)延長(zhǎng)吸附周期,可以在保證產(chǎn)品純度的前提下,降Fig 5 Unit performance us velocity ratio of purge/feed低氣比,節(jié)省操作費(fèi)用(ta=60s,=2s,P2=3,1×l0Pa,p=1.0×10a)O口ep符號(hào)說(shuō)明0.40c——床層氣相濃度,nolm30.35c——床層氣相總濃度,molm-3D—微孔擴(kuò)散系數(shù),cm2s-11.0D——床層軸向擴(kuò)散系數(shù),cm2s-1K——吸附平衝常數(shù)0.20k—?dú)夤虃髻|(zhì)系數(shù),s0.151618202224L—吸附床層長(zhǎng)度,cm吸附塔壓力p——吸附操作階段壓力,PaFig 6 Unit performance ts retaining time均壓操作階段壓力,P(tn=60s,l=2s,P2=3.1×10Pa,Pp=1.0×10Pa)P吹掃操作階段壓力,PaO囗exp.;sinu.吸附相濃度,mom-3吸附平衡濃度molm下,在所考察的范圍內(nèi),適當(dāng)延長(zhǎng)吸附階段的時(shí)微孔半徑,cm間,在產(chǎn)品純度并無(wú)明顯改變的前提下,可以顯著床層溫度,K提高氮?dú)饣厥章蕋-90時(shí)吸均氣間附壓相時(shí)時(shí)流間間0.27速操作條件下吸附階段入口流速-操作條件下吹掃階段入口流速摩爾分?jǐn)?shù)607080Z—量綱1床層長(zhǎng)度z—床層長(zhǎng)度變量,cmFig. 7 Unit performance us operating perica——吹掃/進(jìn)料流速比c床層空隙率OD exp保留時(shí)間——傳質(zhì)系數(shù)乘積因子結(jié)論上角標(biāo)從實(shí)驗(yàn)和模型化兩方面對(duì)SA空分制氮過(guò)程下角分初始狀態(tài)進(jìn)行了較為深入的研究,得出如下結(jié)論A—氧組分590化報(bào)2000年10月1333-1343References4 Hassan MM, Raghavan NS, Ruthven D M. CES, 1987, 428):1 Raghavan N S, Hassan MM, Ruthven D M. AlChE J., 1985, 31(3):385-395 Farooq S, Ruthven D M, Boniface H A. CES, 1989, 44( 122 Raghavan N S, Ruthven D M. AIChE J, 1985, 31( 12): 20172809-28166 Lu Hong Fi ). Experimental Study and Numerical Simulation of Ai3 Hassan MM, Ruthven DM, Raghavan N S CES., 1986, 41(5)Separation by Pressure Swing Adsorption Process: dissertation I i1iex). Beijing: Beijing University of Chemical Techno1998PSA PROCESS FOR NITROGEN PRODUCTION FROM AIRLu Hong and Li ChengyueCollege of Chemical Engineering, Beijing University of Chemical Technology, Beijing 100029)Abstract The PSA( Pressure Swing Adsorption )cycling process for nitrogen production based on the adsorption sepa-ration of air on a commercial carbon molecular sieve was systematically studied. Experimental investigation was carriedout with a pilot plant packed with BF-CMS of 15kg. Transient axial concentration and temperature profiles were measured with a set of seven probes respectively by means of a PC-data gathering system. A mathematical model describing this process was also developed which included a Linear Drive Force mass transfer equation and a Henry adsorptionequilibrium equation. Some of results obtained in a pilot plant mainly product purity and nitrogen recoverymovement of concentration profile in the adsorption bed during adsorption and purge steps ) and comparison between ex-perimental data and results predicted by the model were reported. The reliability of the model was verified and parametricanalysis based on the model was completed. The experimental study and numerical simulation showed that for an existingPSA air separation system a set of specific operating parameters could be found it to obtain optimum performance withthe help of numerical simulation. The model could predict the performance of Psa air separation system perfectlyKeywords air separation, PSA, nitrogenTo whom correspond第51卷第5期盧洪等:變壓吸附空分制氮過(guò)程的研究9附錄變壓吸附空分制氮數(shù)學(xué)模型PSA空分制氮過(guò)程的動(dòng)態(tài)行為可以用如下方程組來(lái)描述(1)流動(dòng)氣相各組分質(zhì)量衡算D0(1)1-)q=0(2)式中(2)氣相總質(zhì)量衡算3)氣固質(zhì)量傳(固相質(zhì)量衡算)kAqa-9A)(5)式中g(shù)A =KACA書(shū)=KBCB=K(c1-cA)A·Dr2·D./r2lg如果不計(jì)二階軸向擴(kuò)散項(xiàng),則式1)~式3沖只有兩個(gè)是獨(dú)立的,例如可以保留式1廂和式3),因而模型微分方程由式1),式3)~式5成(4)邊界條件DL(6)(7)=0=10(加壓吸附時(shí))(降壓均壓時(shí))降壓吹掃時(shí))(8)=(t)(加壓均壓時(shí))y(9)#:指該值為另一塔均壓階段出口處流速5)初始條件潔凈床層z0)=0;q(z0)1(z0)(10)飽和床層cAz 0)=cA: 9(z 0)=KACA(:D)=k=K(e2-g式中z=0指沿流體流動(dòng)方向的床層入口處L指沿流體流動(dòng)方向的床層出口處